Campaigning to the New American Electorate

Advertising to Latino Voters

Marisa Abrajano

• Publisher: Stanford University Press
• Year: 2010
• ISBN:
  9780804774703
• Edition: 1
• Topic: Politics & International Relations
• Subtopic: Political Campaigns & Elections
• Index: Politics & International Relations

Presuming that a strong relationship exists between one's identity and political behavior, American politicians have long targeted immigrant and ethnic communities based on their shared ethnic or racial identity. But to what extent do political campaign messages impact voters' actual decisions and behaviors?

This new book is one of the first to examine and compare the campaign efforts used to target Latinos with those directed at the rest of the electorate. Specifically, it focuses on televised Spanish and English-language advertising developed for the 2000 and 2004 presidential elections, as well as for dozens of congressional and statewide contests from 2000–2004. Author Marisa Abrajano's research reveals exposure to these televised political ads indeed impacts whether Latinos turn out to vote and, if so, for whom they vote. But the effect of these advertising messages is not uniform across the Latino electorate. Abrajano explores the particular factors that affect Latinos' receptivity to political ads and offers key findings for those interested in understanding how to mobilize this critical swing group in American politics.

1

Campaigning to a Changing American Electorate

Hispanics are different to communicate to . . . the Hispanic community is not homogenous.... They are complex in their own and unique ways, less partisan, more independent. . . . I don’t pretend to understand . . . but I do know enough to hire people who do.¹

—Bill Knapp, 2000 Gore campaign media strategist

IN THE SPRING OF 2000, the leaders of the Democratic Party faced a crisis that would have serious repercussions for the upcoming presidential election. The issue involved a boy by the name of Elian Gonzalez who had illegally crossed the Florida straits along with his mother and others from Cuba. Elian’s mother died in the crossing, and the U.S. government was faced with a decision—send Elian back to his father in Cuba or grant him permanent residency in the U.S. so he could reside with his relatives in Florida. Vice President and Democratic presidential candidate Al Gore split with the administration and supported a congressional bill that would provide the boy with permanent residency. Gore adopted this position with the hope of securing the support of Florida’s crucial voting bloc of Latinos, composed mostly of Cuban Americans but also of Central and South Americans. The Clinton administration, however, decided to go in the opposite direction. On April 13, 2000, the Department of Justice sent eight fully armed agents from the Immigration and Naturalization Service (INS) to forcibly remove Gonzalez from his great-uncle’s house and return him to his father in Cuba. The handling of the Elian Gonzalez case proved to be disastrous for Gore’s support among Florida Latinos.² It infuriated the Cuban community and, as scholars Kevin Hill and Dario Moreno (2005) note, Cuban voters used the 2000 presidential election as a referendum on the Clinton-Gore administration’s handling of the affair. In light of the fallout from this incident, the Gore campaign surrendered any hope of winning the Latino vote in Miami and minimized its Spanish-language outreach efforts. This was hailed as a crucial mistake, as noted by Paulo Izquierdo, one of the consultants in charge of Gore’s Spanish-language campaign: “With a true 20-20 hindsight, I’m sure it would have made a big difference . . . We think that if we had been stronger in Miami we would have won Florida. We could have picked up those thousand votes very easily” (Oberfield and Segal 2008, 297).³

Undoubtedly, as this sequence of events reveals, the Latino electorate and Latino outreach efforts play a critical role in the electoral success of any presidential hopeful. Bill Knapp’s sentiments, however, illustrate that what to say to the Latinos and how to say it can be a challenge, particularly in the high-stakes game of presidential campaigns. The Latino population, as defined by the U.S. Census Bureau, encompasses an ethnically and culturally diverse group of individuals from Mexico and Latin America.⁴ People of Latino origin also vary tremendously in their histories and backgrounds. Whereas much of the southwestern portion of the U.S. once belonged to Mexico, many Cubans immigrated to Florida seeking political asylum from the Castro regime; Puerto Ricans, meanwhile, are a unique case because, as American citizens, they can come and go as they please.⁵ Individuals of Latino origin can be fifth- or sixth-generation Americans, recent immigrants, or somewhere in between. Although these are just a few of the features that contribute to the heterogeneity of the Latino population, they exemplify some of the difficulties that politicians face in courting this group of voters.

Latinos’ rising prominence in American politics can primarily be attributed to their rapid demographic growth over the past three decades. Consider, for instance, the following realities—Latinos in 2009 make up 15 percent (44.3 million) of the total U.S. population, but by 2050, the U.S. Census projects that Latinos will constitute a quarter of the population. From 1990 to 2000, the Latino population increased by 57.9 percent, from 22.4 million to 35.3 million, while the rest of the U.S. population grew by only 13.2 percent during this same period (Guzman 2001). The rapid growth of the U.S. Latino population can be attributed to two factors: demographics and immigration trends. With a median age of 26.9, the Latino population is younger, on average, than the rest of the U.S. population, which has a median age of 40.1 (U.S. Census Bureau 2007). Because they tend to be younger, Latinos also tend to have higher fertility rates than the population as a whole. A steady stream of immigrants from Mexico and Central and South America over the past four decades has also contributed to the Latino boom. In 1970, only 19.9 percent of the U.S. Latino population was foreign born; by 2000, almost half of the Latino population (45.5 percent) was born outside of the U.S. (Campbell and Lennon 1999).

These demographic changes have important implications for the current and future state of American politics. First, Latinos are still politically “up for grabs”—they have yet to exhibit stable political and partisan preferences to the same degree as African Americans (Frymer 1999).⁶ Although estimates of the Latino electorate’s support for the Democratic presidential candidate have generally hovered around 60 percent for the past thirty years (Schmal 2004), many more Latinos than Anglos report being independents or uncertain about their partisan preferences (Hajnal and Lee 2008). Forty-seven percent of Latino respondents in the 1993–94 Multi-City Survey of Urban Inequality refused to answer the party identification question, responded that they “didn’t know” which party they supported, or indicated that they had no preference for either of the two main political parties. The 2006 Latino National Survey similarly found that 16.6 percent of Latinos report being independent, 16.3 percent “don’t care” about their party identification, and 20.1 percent don’t know or do not consider themselves to be affiliated with a political party (Abrajano and Alvarez, forthcoming).

Recent elections have also revealed Latinos’ lack of partisan rootedness. In the 1998 midterm elections, Republican candidates were able to secure more than one-third (36.3 percent) of the Latino electorate. And in the 2004 presidential election, the National Election Pool estimates that 53.3 percent of Latinos supported the Democratic candidate, John Kerry, while 44 percent cast their ballots for the Republican incumbent, George W. Bush. This marked the first time that Republicans captured 40 percent or more of the Latino vote in a presidential or midterm election.⁷ Republicans interpreted this level of support as a sign that Latinos’ political preferences are changing, whereas the result convinced Democrats that they cannot assume Latinos will automatically support their candidates.

Much of the Latino population is also geographically concentrated in battleground states and those with a large number of electoral college votes. In each of the five states of California, Texas, Florida, Illinois, and New York, Latinos constitute between 20 and 35 percent of the statewide population; these five states alone account for 168 total electoral college votes, which is more than half of the votes needed for victory in a presidential election. Latinos similarly make up a sizable segment of the statewide population in battleground states such as Arizona, Colorado, Nevada, and New Mexico. And in the politically important southern region of the United States, the Latino population is growing faster than in any other region of the country (Kochhar et al. 2005). For instance, Mecklenburg County, North Carolina, which includes the city of Charlotte, experienced a 500 percent increase in the size of its Latino population from 1990 to 2000 (Kochhar et al. 2005). Latinos are important political players, then, not only in the states in which they have traditionally settled (primarily in the southwestern United States) but also in nontraditional settlement areas across the nation. The geographic concentration of Latinos in key states, along with the instability of their political orientation, heightens their attractiveness to candidates and political parties. For these reasons, the Latino electorate has been considered the critical swing group and the “sleeping giant” of American politics.

But as Bill Knapp’s chapter-opening quote suggests, the Latino population, when considered as a pool of potential voters, is substantially different from the national population as a whole. Primarily because forty percent of the Latino population is born outside the U.S. (U.S. Census Bureau 2007), their familiarity with American politics is not the same as it is for the native-born population. How these immigrants conceptualize basic political terms, such as liberal and conservative or Democrats and Republicans, is likely to vary from native-born Americans, given that these terms are unique to the U.S. political system. This unfamiliarity is reflected in survey responses from the 2006 Latino National Survey (Fraga et al. 2006), in which nearly one-third (31.6 percent) of Latinos did not consider themselves liberal, moderate, or conservative. In addition, the concept of multiparty politics is relatively new to some Latinos (Kasinitz et al. 2008). Take, for example, the case of Latinos of Mexican heritage: a one-party system led by the Institutional Revolutionary Party (PRI) ruled the country from 1929 to 2000 (Cothran 1994). The children of immigrants are also affected by this lack of familiarity because children learn about politics and acquire partisanship mainly through their parents (Vaillancourt 1973). The Latino population also differs from the rest of the U.S. population on indicators pertaining to socioeconomic well being. According to U.S. Census estimates, Latinos’ median household income is lower than that of Anglos, Asians, and African Americans, and only about 60 percent of Latinos graduate from high school—the lowest graduation rate of any racial or ethnic group in the country (U.S. Census Bureau 2007). Approximately one out of five Latinos lives in poverty, and almost one out of three (32.7 percent) lack health insurance (U.S. Census Bureau 2007).

In light of the fact that socioeconomics and familiarity with politics are strong predictors of political participation (Verba et al. 1995), Latinos have yet to reach their full political potential. While their total share of the U.S. population stands at 15 percent (U.S. Census Bureau 2007), Latinos make up 6 percent of the total American electorate. In 2004, about 9.3 million Latinos—less than a quarter of the total U.S. Latino population—registered to vote, with 7.6 million having reported voting. As a point of comparison, almost 16 million African Americans were registered in 2004, and 14 million reported voting (Abrajano and Alvarez, forthcoming). Thus, although Latinos constitute a larger share of the total U.S. population than do African Americans, they make up a smaller percentage of voters. This difference is partly because a significant number of Latinos face another hurdle before registering to vote—that of obtaining citizenship. The Current Population Survey estimates that only 63 percent of Latinos of voting age are citizens (U.S. Census Bureau 2009).⁸

Still, the growing size and influence of the Latino electorate presents a compelling opportunity for politicians to win over new voters and, ideally, lock in the allegiance of this group in the hope of future returns. Yet the particular demographic and political characteristics of the Latino electorate offer challenges to politicians as well, because it is not yet clear which strategies of campaigning are likely to win these voters over, and it is unclear whether Latino voters will behave like other, well-understood segments of the electorate. Conventional wisdom holds that candidates will advertise differently to ethnic and racial groups than they will to Anglos, because candidates are targeting a specific group of voters who presumably share something in common (Erie 1990; Popkin 1994; DeFrancesco Soto and Merolla 2008; Shea and Burton 2006). Among Latinos, that commonality is based on a shared language and, in some cases, a common cultural heritage. One can conceive of a group in much the same way that Bishin’s work on subconstituency politics does, as “a constellation of people, either organized or not, who share a social identity owing to a common experience that leads to shared concerns and preferences” (2009, 5). Appealing to individuals based on culture is a powerful and relatively easy way to show voters that candidates understand and can relate to them (Popkin 1994). Indeed, politicians used ethnic campaigns on European immigrants who arrived in the U.S. in the late 1800s (Erie 1990; McNickle 1993). Presumably, then, these campaigns work, because politicians have been using them for decades. So we should expect nothing less of aspiring politicians today; when campaigning to our nation’s largest immigrant group, they too should adopt a distinct ethnic political campaign.

The aim of this book is twofold. My first goal is to determine whether ethnic political campaigns are successful at winning ethnic minority votes. I challenge the conventional wisdom that all minorities will react to these campaign appeals in a positive way. The responsiveness of a given ethnic minority group to these political messages will vary based on the extent to which they have incorporated into American political life. My second goal is to examine the consequences, if any, that ethnic political campaigns have on the political health and well being of a segment of the ethnic group being targeted.

In addressing these issues, this book focuses specifically on the Spanish-and English-language televised political ads created for the 2000–2004 election cycles and examines the effects of the ads on Latino political behavior. Does exposure to these political ads influence who they vote for, whether or not they vote, and their knowledge of the candidates? Indeed, variations are likely to exist in the content of Spanish- and English-language ads, so that ads produced in Spanish may emphasize personal and non-policy-based appeals to a greater extent than do English-language ads. The reason for these distinctions, which serves as the basis for the theory of information based–advertising, is due to candidates’ perceptions of Latinos’ orientation to politics and to candidates’ beliefs about the importance of ethnic identity as a factor in Latinos’ political decision making. In general, because a large portion of the Latino population is relatively new to American political life, the campaign messages used to target this group of voters is fairly simple and symbolic in nature, focusing more on cultural cues and references than on candidates’ policies and issue positions. The effectiveness of these campaign messages varies based on a Latino’s familiarity with and knowledge of American politics and on the salience of ethnic identity in one’s political behavior.

The research endeavor undertaken for this book is important for the well being of our nation’s latest newcomers. First, if candidates are indeed advertising in the manner just described, it raises the possibility that the parties and candidates are acting in ways that systematically disadvantage some segments of the population in becoming informed participants in the political process. Because televised political ads aid in citizen learning by increasing one’s knowledge of politics (Jackman and Vavreck 2009; Geer 2006; Patterson and McClure 1976; Brians and Wattenberg 1996), ads that mostly emphasize personal or cultural appeals make it difficult for individuals to learn anything about the candidates’ policy positions. Adopting this type of advertising strategy also weakens the ability of certain communities to participate in politics the same way that others do, and, in turn, to have their voices heard. Finally, a bias in advertising could serve to undermine a basic principle of democracy—that all people are treated equally under the government. Several of these consequences are investigated in this book.

The Power of Televised Political Advertising

In attempting to communicate with and win over voters, politicians have innumerable tools at their disposal, including campaign appearances, direct mailings, websites, televised debates and speeches, and door-to-door canvassing, to name just a few. But in the era of “new style” campaigning (Agranoff 1972), televised political advertisements have become the primary means by which candidates communicate with voters. This is clearly reflected in the way candidates allocate campaign funds. In the 2000 election cycle, approximately 85 percent of candidates’ advertising budgets went to broadcast “spot” television advertisements (Segal 2002). The amount of money spent on advertising is staggering: in 2000, presidential candidate George W. Bush spent $75.3 million on televised campaign ads, while his opponent, Al Gore, allocated $77.1 million.

It is also by means of televised advertisements that candidates have most notably adjusted their behavior to take advantage of the growing influence of Latino voters. In the 2000 election cycle, both George W. Bush and Al Gore devoted approximately $4 million to Spanish-language television commercials. This spending marked the largest amount ever allocated by presidential candidates to Spanish-language advertising, and the millions of dollars devoted to appeals in Spanish has steadily increased since (Segal 2002).

That candidates spend such vast sums on television commercials is an acknowledgment of these advertisements’ power to influence voters. The advent of televised political advertising in the presidential election of 1948 initially raised great concern about this power among scholars, the media, and the general public. This fear led to popular claims that political ads were deceitful and overly personalistic and lacked substantive material (West 2001, 44, 46). Scholars and the general public believed that, in advertisements, “the candidate’s personality, image, and symbolic appeals [take] precedence over specific issue positions” (Atkin et al. 1973, 210). Berelson and colleagues (1954) alleviated some of these concerns; their research concluded that short-term effects, such as campaigns, had a minimal influence on voter decision making. Berelson and his colleagues argued that political campaigns did not provide new information to voters but instead triggered voters’ previous beliefs and opinions. These findings led many scholars to support the “minimal effects hypothesis,” which contends that campaigns have little or no impact on a voter’s decision-making process.

Decades of research following Berelson’s seminal work, along with technological advancements in measuring advertising exposure, leave little doubt that television ads influence voters’ opinions and attitudes in presidential elections and nonpresidential elections alike (Brader 2006; Geer 2006; West 2001; Mendelberg 2001; Just et al. 1996; Herrnson and Patterson 2000; Kahn and Kenney 1999). As one might expect, more frequent advertising positively a...

Table of contents

• Title Page
• Copyright Page
• Dedication
• Table of Contents
• Acknowledgments
• 1 - Campaigning to a Changing American Electorate
• 2 - A Theory of Information-Based Advertising
• 3 - Campaigning to Ethnic and Racial Minorities in the U.S.
• 4 - Candidates’ Advertising Strategies
• 5 - Advertising Effects on the Latino Vote
• 6 - The Consequences of an Information-Based Advertising Strategy
• 7 - The Future of Ethnically Targeted Campaigns
• 8 - Epilogue: The 2008 Campaigns
• Appendix A: Coding the Advertisements
• Appendix B: Constructing the Ad Exposure Variable
• Appendix C: Voter Learning and Vote-Choice Model Specification
• Notes
• References
• Index