Editor’s Note: We are proud to publish today an original article by two noted academic experts on the highly relevant topic of Hispanic voters in 2008. The authors are R. Michael Alvarez, a professor of political science at Caltech in Pasadena, and Jonathan Nagler, a professor of politics in the Wilf Family Department of Politics at NYU. Together they have studied voting behavior in recent presidential elections, and have written a number of papers on Hispanic political behavior. In 2004 they were involved in Hispanic research for the Kerry campaign, and have worked on a number of Hispanic research projects in association with Greenberg, Quinlan and Rosner Research.
We also anticipate recieving and publishing some comments and rejoinders from other experts in this field over the next couple of weeks, and intend to continue this discussion until election day and beyond.
Winning the Hispanic Vote in 2008
by R. Michael Alvarez and Jonathan Nagler
Introduction
Historically, Democratic presidential candidates have done quite well with Hispanic voters (with some exceptions, such as Cuban-Americans). For the past three decades, Democratic presidential candidates have typically received more than 60% of the votes cast by Hispanics.
But in the 2004 presidential election, Hispanic support for John Kerry was lower than the historic norm. While there has been much debate over the exact percentage of support that Kerry received from Hispanic voters in 2004, a consensus has emerged that at best Kerry might have received 60% of the Hispanic vote. But no matter what we think the exact percentage was, Hispanic voters were attracted to Bush in greater percentages in 2004 than to any previous Republican presidential candidate in recent history (See David L. Leal, Matt A. Barreto, Jongho Lee, and Rodolfo O. de la Garza, “The Latino Vote in the 2004 Election”. PS: Political Science and Politics, v. 38, 41-49, 2005; Marisa A. Abrajano, R. Michael Alvarez and Jonathan Nagler, “The Hispanic Vote in the 2004 Presidential Election: Insecurity and Moral Concerns”, Journal of Politics, forthcoming (April, 2008); David L. Leal, Stephen A. Nuno, Jongho Lee, and Rodolfo O. de la Garza, “Latinos, Immigration, and the 2006 Midterm Elections.” PS: Political Science and Politics, v. 41, 309-317, 2008.).
There are two questions that Kerry’s performance with Hispanic voters in the 2004 presidential election raises. One question is why — what was it about the context of the 2004 presidential election, and the messages articulated by Kerry and Bush, that caused more Hispanics to support Bush than is normal for a Republican presidential candidate? The second question is what does this imply for the 2008 presidential election — what strategies should the Barak Obama, the presumptive Democrat nominee, pursue to insure a stronger performance among Hispanic voters in November 2008?
In this article we provide answers for both of these questions.
What Happened in 2004?
The Kerry campaign appeared to treat the Hispanic vote seriously in the 2004 election. Just prior to the Democratic National Convention in late July 2004, the Kerry campaign announced an unprecedented financial investment aimed at targeting Hispanic and African-American voters. At that point in the 2004 campaign, the stage seemed to be set for Kerry to devise a strong appeal to Hispanic voters.
But that appeal failed to take into consideration the potential Republican election strategy, as well as the particular context of the 2004 presidential race. In research we have done with Marisa A. Abrajano (The Journal of Politics, 2008), we found that despite their concerns about the national economy and the war in Iraq, Hispanic voters were attracted to Bush because of two appeals: first, his stance on moral values; and second, his national security message.
In our paper, “The Hispanic Vote in the 2004 Presidential Election: Insecurity and Moral Concerns”, we used exit poll data from most of the states with large Hispanic electorates to develop a statistical model to determine the issues that motivated Hispanic voters to support Kerry or Bush (We used respondents from Arizona, California, Colorado, Florida, Nevada, New Mexico, New York, and New Jersey. These were the states where the Hispanic population was at least 6% of the state population, and where the necessary questions were asked on the Exit Poll. In Connecticut, Illinois, Massachusetts, and Texas respondents were not asked what they felt the most important issue was.)
We then used our model to examine support for Kerry or Bush in two different hypothetical scenarios. In the first scenario we simulated one aspect of campaign strategy: what if Kerry had been successful in completely neutralizing particular issues, that is – convincing all Hispanic voters that the issue was not important? In the second scenario we considered what would have happened if instead of completely neutralizing issues, one of the candidates had persuaded all Hispanic voters that a particular issue was the most important issue of the election to them? So for example, in scenario one, what if Kerry had successfully persuaded all Hispanic voters that moral values were not a concern, or that national security issues were not a concern? Or, for scenario two, what if Kerry had convinced all Hispanic voters that education was the most important concern for the election, or that health care was the most important concern for the election?
This analysis revealed that had Kerry managed to neutralize select issues, two issues would have been powerful in moving Hispanic votes into his column: terrorism and moral values. Our statistical model predicts that if had Kerry completely neutralized the moral values issue, his vote share among Hispanics in the states we analyze would have increased 2.2 percentage points, from 60% to 62.2%; had he done the same with terrorism, his vote share among Hispanics would have increased 2.7 percentage points, from 60% to 62.7%. As Kerry lost the popular vote by less than 2.5 percentage points, obviously these are meaningful swings in the vote.
Compared to moral values and terrorism, traditional Democratic issues such as education and health care played relatively little role in 2004. Performing similar counterfactual analyses to those we describe above, our model predicted that had no Hispanic voters felt that education or health care was the most important issue, then Kerry’s vote share would have dropped by only 0.5 percentage points and 0.6 percentage points respectively (from 60.0% to 59.5% and 59.4%, respectively).
This does not bode well for a Democratic candidate: the alternative way to frame this is that Kerry only convinced enough Hispanic voters to believe that education or health care were the most important issues in the campaign to raise his vote share less than one percentage point compared to an electorate where no Hispanics thought either of these were the most important issue. Yet these are issues that Hispanics have traditionally claimed were
important. This suggests Kerry was simply not winning enough votes in an issue area that has traditionally favored Democrats. Our model predicts that if Kerry had done well in this area and convinced all Hispanic voters that education or health care was the most important issue, then his vote share would have risen by 5.3 or 11.6 percentage points, respectively. Obviously convincing all Hispanic voters that either of these was the most
important issue was not feasible. But in the states we examined, only 8.4% of Hispanics listed education as the most important issue, and only 6.8% of Hispanics listed health care as the most important issue. These percentages are well below the percentages of Hispanics who have listed these as major concerns in polling conducted prior to the election contest.
What Does This Mean For 2008?
Clearly, much has changed since the 2004 presidential election. The second Bush term, the 2006 Democratic successes in the midterm elections, the continued war in Iraq, turmoil in the housing market, rising prices, and signs of economic recession all will help to shape the context of the 2008 general election. And since 2004 the issue of illegal immigration has also risen in national concern; recent surveys of Hispanic voters show immigration and in particular efforts to deal with illegal immigration to be an important concern. Recent polling has shown the immigration issue to be an important one for other voters as well, especially white voters across the nation.
Given the prominence of the immigration issue, especially for Hispanics, and the failure of federal efforts to devise legislative solutions to help resolve the problem of illegal immigration, many have argued that the immigration issue might be one that the Obama and the Democratic Party could use to their advantage in the 2008 presidential race. But now that the context of the presidential race is becoming clearer, it is no longer the case that the immigration issue will necessarily be important in the 2008 general election, nor an issue that the Obama can easily use to win Hispanic votes. There is one reason for this — the presumptive Republican nominee, John McCain.
McCain has in the past has backed comprehensive immigration reform, most recently in the Senate bill he cosponsored with Ted Kennedy that would have, among other reforms, created an “essential worker visa program.” While risking the possibility that he will anger the more conservative elements in the Republican party that desire strong action on immigration (such as building a wall along the US-Mexican border, etc), McCain’s past stance on immigration will make it difficult for Obama to easily draw clear distinctions on this issue, and may effectively reduce the prominence of the immigration issue in the general election.
If that happens, we believe that McCain will draw directly from the 2004 Republican playbook when it comes to the Hispanic electorate. He is likely to turn again to national security (an issue where his background as a Vietnam veteran, his support for the surge in Iraq, and his legislative career in Congress give him strong credibility) and moral values as issues in his messaging to Hispanic voters. As in the 2004 election, we believe that this may again erode Hispanic support for Obama. But there are ways in which the Democrats can develop messages that can mitigate, if not eliminate, the potential threat which moral values and national security issues may pose in November 2008.
Consider the current context: continued fallout from the sub-prime mortgage crisis; plummeting housing values, soaring numbers of foreclosures, and a lack of credit for purchasing homes. Basic costs of living for food and especially gasoline are rising dramatically, with $4 per gallon gas common across the country. We have clear signs of an ongoing economic slowdown, if not recession. Consumer confidence is sagging. Many families, including many Hispanic families, cannot afford adequate health care, and in many instances have no health care coverage at all. The American casualty count in Iraq has passed 4,000, and there are no signs that the Iraq War will end soon.
Data from Hispanics who participated in the recent Super Tuesday primaries, as compiled by the Pew Hispanic Center, shows that this basic trilogy of issues are of importance to Hispanic Democratic voters: the economy (53%), the war in Iraq (24%), and health care (21%). Importantly, both white and African-American Democratic voters on Super Tuesday perceived these same three to be of importance, in the same relative order; the economy, the Iraq War, and health care.
Thus, there is a basic narrative that Democrats can – and must – develop in order to have a strong Hispanic strategy in 2008. That narrative needs to focus on the core strengths that the Democratic candidate will bring to the table: a progressive message that articulates how the federal government will bring the nation economic growth, how it will provide affordable housing and credit to middle and lower income families, how it will deal with skyrocketing costs for food and energy, how it will make health care available and affordable, how it will make quality education a priority, how it can bring high-quality jobs to all those who want them, and how they will end the war in Iraq.
We think that for Democrats to retake the White House in 2008, they must work to get strong Hispanic support — and keep Republican Hispanic support at or below the 35% threshold. Doing this will not be easy, but will require that the Democrats take advantage of the constellation of domestic issues that work so strongly in their favor with Hispanic voters: the economy, education and health care.