Colin Powell's warning to George W. Bush before the invasion of Iraq, that "this will become the first term," proved overly optimistic (Woodward 2004, 150). The war consumed Bush's second term as well, leaving a radically altered political configuration in its wake. The 2004 election had elevated Bush's Republican Party to its most dominant position in national government since the 1920s. Four years later, Barack Obama, a young African American Democrat unknown to most Americans at his campaign's outset, occupied the White House, and Democrats controlled the House and Senate by margins as wide as before the Republican revolution of 1994. The war was not the only reason for this remarkable reversal of party fortunes; the economy's downward slide during 2008 and then plunge off the cliff in September contributed to it, as did other Bush administration failures, such as the response to Hurricane Katrina. But the economic crisis was so devastating because it compounded damage that the war had already inflicted on the president and his party. Although the situation in Iraq improved markedly over the year leading up to the election, a development for which Bush and his controversial "surge" strategy could take considerable credit, the damage proved irreversible. Though the economy became the dominant electoral issue in 2008, the Iraq War was, through direct and indirect pathways, ultimately the single most important contributor to Obama's presidential victory.
My purpose here is to flesh out and defend this argument empirically. I begin by describing the political context that had emerged by the end of the Bush presidency, largely as a consequence of the public's reaction to the war and the president. I then examine how the war and the president influenced the primary and general election outcomes, mainly through analyses of two major election studies, the time-series component of the 2008 American National Election Study (ANES 2009) and the 2008 Cooperative Congressional Election Study (CCES; Ansolabehere 2009a). Finally, I briefly compare Bush's effect on his party's nominee to the influence of Ronald Reagan and Bill Clinton on the fates of George H. W. Bush and Al Gore.
The War, the President, and the Republican Party
The most important effects of the Iraq War on the 2008 presidential election had already registered well before the first caucuses in Iowa. Because I have already presented the evidence for this conclusion in some detail (Jacobson 2009b, 2009c), I merely summarize it here. Popular support for the Iraq War and approval of Bush's performance as president bore close to a one-to-one relationship from the war's initiation until early 2008. As Americans soured on the war, they soured on its architect as well. The monthly average of support for the war measured in national surveys fell from 73% in April 2003 to 33% five years later. (1) Bush's mean monthly approval ratings fell in near lockstep, dropping from 71% to 28%. The close correspondence between the two trends was no accident, for the proportion of respondents in individual surveys who offered consistent responses to questions about the president and the war--positive toward both or negative toward both--remained steady at an average of about 84% across the entire period (Jacobson 2009b, 175).
As the war and the president grew increasingly unpopular, so did the Republican Party. The mean proportion of Americans rating it favorably fell from 55% in surveys taken during the first half of 2003 to 41% during the second half of 2008; meanwhile, the Democratic Party's favorability ratings grew modestly, from 52% to 55%, leaving the Democrats with a 14-point advantage on this dimension, among the widest ever recorded (Jacobson 2009c, 8). (2) More crucially, the Republicans' share of major party identifiers also shrank over this period, dropping by nearly 7 percentage points and leaving the party at its lowest ebb since the early 1980s (Jacobson 2009b, 194-95). …