How Does Obama Match-Up? Counterfactuals & the Role of Obama’s Race in 2008

Would the outcome of the 2008 election have been different if Hillary Clinton, John Edwards, or a past Democratic nominee (like John Kerry or Al Gore) had been the nominee? In this paper we treat the counterfactual model earnestly and compare 33 different presidential elections, some real and some hypothetical, in order to understand how candidate traits and election-level characteristics affect outcomes. We do this paying particular attention to the role of racial prejudice and whether it helped or hurt Barack Obama at the polls. Our findings suggest that “old-fashioned” racial stereotyping was singly important in decisions about Obama in 2008, relative to its role in past elections or in 2008 choices substituting Clinton or Edwards for Obama. Its impact, however, was steady over the course of the campaign relative to the dynamic effect of “new” or symbolic racism, which gained in importance over the year prior to the election and was also exceptionally important when Obama was in the choiceset. Obama balances the losses due to racial prejudice with gains from cross-partisan appeal, doing better than Clinton or Edwards (and better than most past Democratic candidates) at attracting Republicans and independents. The Democrats would have won the 2008 election regardless of who they nominated, but that the average Democratic party nominee from the last 16 years, and either of Edwards or Clinton, would have done better against McCain than Obama, although in Clinton’s case, not by much. Contrary to the claims of some of my critics . . . I have never been so näıve as to believe that we can get beyond our racial divisions in a single election cycle, or with a single candidacy – particularly a candidacy as imperfect as my own. – Barack Obama, March 18, 2008 On March 18th, 2008, Barack Obama asked Americans to engage in a conversation with him about race in America — and specifically, not to sweep the events of the last few days under the rug and “hope they will fade away.” Obama was leading the Democratic race to be the party’s presidential nominee by only 148 pledged delegates. The Pennsylvania primary was right around the corner, and his Pastor, Reverand Jeremiah Wright, was becoming famous for his incendiary language about the endemic nature of white racism in America. As hard as he tried, Obama could not separate his message of hope and unity from his opponents’ signals about race and the media elite’s frame of polarization and division. Despite Obama’s vision of a “colorblind America,” we argue in this paper that voters’ attitudes about race played a critical role in the 2008 presidential election, structuring and defining political behavior with unprecedented magnitude and a heretofore unappreciated dynamic over the course of the campaign. Relative to previous Democratic party nominees and others who ran for the nomination in 2008, the effects of attitudes about race, both old-fashioned, stereotypical views and more subtle symbolic prejudices, are exaggerated in 2008 when Obama is the Democrat on offer. Despite this Obama exceptionalism, we conclude that a Clinton nomination would not have resulted in a dramatically different share of the two-party vote (among whites) since Obama attracted many independent and Republican voters who Clinton could not draw to the party. John Edwards, however, or the average Democratic party nominee (since 1992), would have On January 26th just after Obama won the South Carolina primary, a reporter asked former President Bill Clinton, who was helping his wife Hillary campaign for the nomination, “What does it say about Barack Obama that it takes two of you to beat him?” Clinton, clearly bemused by the question, responded by saying “Jesse Jackson won South Carolina twice in ’84 and ’88; and, he ran a good campaign, and so has Obama ran [sic] a good campaign.” In the analyses that follow, we hold the set of voters who make up the electorate constant. See Pasek et al (2010, p. 981) for a different approach. Their mobilization analyses conclude, however, that “switching candidates was much more common than moving into or out of the participating electorate” in the 2008 general election.