We are experiencing COVID-related supply chain delays. Please note, orders are currently taking 10-15 days to be delivered.
We thank you for your understanding and patience.
Penn Press logo
Electing the President, 2004

In this volume, the consultants who brought the country the Bush-Cheney and Kerry-Edwards presidential campaigns of 2004 explain the strategies behind the ads and debates, discuss what they did and failed to do to elect their candidates, and reveal their differing perspectives on the issues that mattered.

Electing the President, 2004
The Insiders' View

Edited by Kathleen Hall Jamieson

2006 | 264 pages | Paper $27.50
Political Science / American History
View main book page

Table of Contents

Campaign Timeline and Charts

1. Campaign Organization and Strategy
2. Advertising
3. Polling: Decisive Moments and Audiences
4. Debate Strategy and Effects
5. The Press/Campaign Relationship
6. Republican Spenders
7. Democratic Spenders

List of Participants

Excerpt [uncorrected, not for citation]


Beginning in 1992, the Annenberg School for Communication has held election debriefings after each general election presidential campaign. First in tape form and, starting in 2000, in transcript form as well, we have shared this event with the scholarly community.

We open with a timeline of election year events and with charts drawn from the data of the National Annenberg Election Survey. These materials were prepared for and distributed at the debriefing.

In some sense the 2004 election was a rematch between key consultants on each side, with Bob Shrum and Bill Knapp among those returning for the Democrats and Matthew Dowd, Mark McKinnon, and Alex Castellanos returning for the Republicans.

As the program outline indicates, the first day of the debriefing, held at the school in Philadelphia, focused on the Bush-Cheney and Kerry-Edwards campaigns. A brief biography of each presenter appears in the text before that person's first presentation.

After a weekend break, the second day, which focused on the major 527s, was held at the National Press Club in Washington, D.C. The consultants who attended the first event did not attend the Washington one, nor did the 527 representatives attend the event in Philadelphia. We had hoped to create a debriefing that happened across two consecutive days in one location, but schedules did not permit us to do so.

As in the past, we asked the consultants on each side to prepare half-hour presentations. We then opened the floor to questions. In Philadelphia, where the event was closed to the press, the questions were asked primarily by the consultants in the audience. In Washington, the questions came largely from reporters. That half-day was carried by C-SPAN.

The goal of the debriefing was to capture the insights of these individuals for examination by scholars and students in the coming months and years. After we transcribed the presentations and discussion, I removed some of the peculiarities of oral speech and added punctuation and bracketed information for context. We then sent the edited version to the participants for correction of spelling and transcription errors. (For example, in the process "youth" had become "use" in one sentence.)

We are grateful to the consultants who, election cycle after election cycle, have helped the scholarly community better understand their take on what happened. I am particularly indebted to Annenberg Public Policy Center staff members Joyce Garczynski and Jennifer Ernst for superintending the events in Philadelphia (thank you, Joyce) and Washington (thank you, Jennifer) and for shepherding the edited transcripts back and forth to our guests. Brooks Jackson, who heads APPC's FactCheck.org, moderated the Washington Debriefing. Annenberg doctoral student and senior National Annenberg Election Survey analyst Kate Kenski moderated the day in Philadelphia. Kyle Cassidy took the photos for the book. The audiences for both days included Annenberg School and Penn political science faculty, APPC staff, and graduate and undergraduate students from Annenberg classes, along with a few special friends of the school and policy center. These included Richard Johnston of the University of British Columbia and Michael Hagen from Temple University, each of whom played an indispensable role in creating the National Annenberg Election Survey, and Margaret and Hank Kenski, who, like Richard and Michael, are distinguished political scientists, but who, unlike Richard and Michael, are Kate Kenski's parents.

Kathleen Hall Jamieson
Elizabeth Ware Packard Professor of Communication
Annenberg School for Communication and
Annenberg Public Policy Center
University of Pennsylvania


In the 2004 post-election period, the Democratic consultants cast the election as close—theirs but for fifty thousand or so votes gone the other way in Ohio. The Republicans cast it as an election that was theirs to lose. They anticipated and got a two- to three-point victory in a contest that, as both sides anticipated, came down to the outcome in Ohio. In larger terms, in the heady post-election days the Republicans cast the election as a mandate for the Bush agenda of Social Security reform, tax simplification, tort reform, and the like.

When all the votes were in, the numbers, mercifully, showed that unlike 2000 the same person had carried both the electoral and popular vote with the final Electoral College spread 286-251.

It had been sixteen years since a presidential candidate won a majority of the national popular vote. The final vote in the November 2004 presidential election gave incumbent president George W. Bush 60,693,281 votes to Senator John Kerry's 57,355,978, a 2-3 percent margin of victory and popular vote majority for the Republicans. Unlike the situation in 2000, independent candidate Ralph Nader played no role in the outcome.

Importantly, as one assesses the effects of the election on governance, 2004 elected Republicans beyond the presidential ticket. The Texas son of a former president was the first Republican incumbent since Calvin Coolidge to gain a second term and net seats in the House and in the Senate at the same time.

This was an election in wartime. Five times in the history of the U.S. a wartime president had sought reelection and five times the wartime president has been reelected. Was the outcome decisively shaped by September 11 or more specifically the public's willingness to amalgamate the war in Iraq to the war on terror? Senator John Kerry thought so. On Meet the Press, January 30, 2005, he told moderator Tim Russert that "9/11 was the central deciding issue of this race." As the debriefing attests, the Republicans worked persistently to reinforce the public perception that September 11 was the decisive event of the Bush first term and also to drive the public's inference that Senator Kerry wasn't up to the job.

The 2004 election was shaped by geography as well. The last non-Southern Democrat to win the presidency was Massachusetts Senator John Kennedy. In 2004, the only Southern state in which the Democrat was competitive was Florida, and there he fell short of winning.

Did the economy matter? In Florida, where the economy was doing well, Bush pulled comfortably ahead of Kerry and outstripped his own 2000 performance. But in Ohio, where job losses were comparatively high, both sides thought Kerry had a chance.

Could Senator John Kerry have won? The deficit was at a record high. More than half the country reported as the election neared that the country was on the wrong track. Presidential approval hovered around 50 percent. The stock market had lost value on Bush's watch and as Election Day approached it looked likely that the incumbent would, as the Democrats were fond of noting, be the first president to lose net payroll jobs since Herbert Hoover. Importantly, as well, the war in Iraq had not gone as forecast, with its costs and casualties higher and its end not clearly in sight on Election Day.

On the plus side for the Republicans was the fact that by a number of key indicators the economy was improving. Indeed the standard political science models forecast a Bush victory. And the turnover of authority to the provisional government in Iraq in late June moved headlines damaging to Bush that before had dominated the front page onto the inside pages of the nation's newspapers.

The 2004 election turned some conventional wisdom in its head. High turnout is supposed to benefit the Democrats. However, in 2004 a precinct-level turnout machine staffed by volunteers gave the Republicans a turnout advantage. Importantly, for the first time in recent memory the number of self identified Republicans who voted equaled the number of people saying they were Democrats. As Democratic pollster Mark Mellman observed, "In the 70s and 80s, Democrats on Election Day had 15-point margins. By the time we got to the 80s, those were 2- and 3- and 4-point margins. It is right to say that today this was the first election where the exit polls showed parity."

In the presidential race, voter turnout was up substantially from 2000. Bush won by about 10.2 million more votes than he had in 2000. Still, Kerry could claim more ballots than any other Democratic contender in history, about 6.4 million more than Gore gathered in 2000. In Ohio, a state in the Republican column in every major race in which a Republican has won the presidency, both parties increased the number of votes they gained over 2000. "Bush countered Kerry's gains in the metropolitan precincts by boosting his margin in exurban and rural counties from 57 to 60 percent, eking out a 118,457-vote victory," Thomas B. Edsall and James V. Grimaldi reported in the Washington Post.

The Bush campaign also took on conventional wisdom by risking attacking the Democrat early and often while running relatively few ads burnishing Bush's image and credentials. Before the nomination was even officially Kerry's, the Republicans had bashed him in over $60 million dollars of ads, giving lie to the common belief that high levels of attack will fatally damage the attacker.

After a decade of forecasts that the upcoming election would be the one in which the Internet would become a force to be reckoned with, both sides agree that 2004 was the election in which it did. Among its functions, in 2004 the Internet served as a fundraising vehicle, a mobilizing tool, an informational channel, and a way for bloggers to make their views known and their insights felt. It also enhanced the impact of the 527s that built webpages around their ads and used the web to appeal for money to stay on the air.

Whether communication mattered is always open to debate but the unprecedented level of paid communication was not in dispute. During election season more than 630,000 ads aired.

However, contrary to expectations that Republicans will outspend Democrats, one side didn't decisively win the race for dollars. In the end, reported Edsall and Grimaldi, "John F. Kerry and his Democratic supporters nearly matched President Bush and the Republicans, who outspent them by just $60 million, $1.14 billion to $1.08 billion."

How that money was deployed may however have made a difference. Edsall and Grimaldi conclude, " Democrats simply did not spend their money as effectively as Bush. ... In a $2.2 billion election, two relatively small expenditures by Bush and his allies stand out for their impact: the $546,000 ad buy by Swift Boat Veterans for Truth and the Bush campaign's $3.25 million contract with the firm TargetPoint Consulting. The first portrayed Kerry in unrelentingly negative terms, permanently damaging him, while the second produced dramatic innovations in direct mail and voter technology, enabling Bush to identify and target potential voters with pinpoint precision."

The campaign by the so-called Swift Boat Veterans for Truth will undoubtedly elicit intense scholarly scrutiny in the coming years. Because of the rules governing when a candidate received federal matching funds and with them a cap on how much he could spend, Kerry fought a general election with fewer dollars to spend per day. In what may have been a strategic error, his campaign conserved its funds in August while it was hammered a small media buy by the so-called Swift Boat Veterans for Truth. Arguably that first half million dollars or so SwiftVet ad buy, amplified by free airing on cable talk shows, was the most effective small dollar ad purchase since the national Security Political Action Committee added "Willie Horton" to the political lexicon in 1988.

The 527s cut both ways in 2004, however. The pro-Kerry 527s may have thwarted the Republicans' effort to derail the Kerry candidacy in the post-primary season. Days after Super Tuesday locked up the Democratic nomination for Massachusetts Senator John Kerry, the Bush team was on the air with a $40 million six-week assault. Since the Kerry camp was effectively out of money, the pro-Kerry 527s moved into the gap with anti-Bush messages. Were it not for their efforts, described in the debriefing by representatives from The Media Fund and MoveOn, the Bush effort might have effectively ended Kerry's prospects before he was even nominated.

In the category of "what might have been" are many questions as well as significant disputes about the answers. Should Kerry have accepted the $75 million in Federal funding and the attendant spending caps that came with it when he accepted the nomination a month before George Bush did?

Is a Senator with decades of votes on the record always a more vulnerable candidate than a governor or incumbent president?

Why did Kerry received a smaller bounce from his convention than the Republican consultants had expected and the Democrats had hoped?

What if any was the importance of the ballot initiatives in 11 states that defined marriage as the union of a man and a woman?

Did the exit poll that showed moral values as a prime motivator for Bush votes get it wrong?

Did the Democrats suffer an insurmountable strategic disadvantage? In place for President Bush with years of planning time on their hands was the campaign team that had eked out a victory in 2000 and now had the added advantage of plenty of time to find the votes that would give their candidate both an electoral and popular vote victory. The Republicans were blessed as well by the fact that the incumbent was unchallenged in the primaries. Moreover, within his own party, Bush was popular. Republicans traditionally have high loyalty from their party supporters.

The Republicans began running for reelection in 2004 before Governor Bush had been inaugurated in 2000. In the weeks after his election in 2000, his representatives buffered their standard-bearer from bad economic news. George W Bush has inherited a recession from Clinton, they argued even before he had been sworn in. While the Democrats and the economic indicators said it was not a recession until months into the first Bush term, no one seriously disputed that the downturn had begun before he was in office. After September 11, the Bush team pinned the faltering economy on the terrorist attack.

Did Kerry focus too intensely on his time in Vietnam during the Democratic convention? The Democrats argue that only 6 percent of his acceptance speech focused on Vietnam. The Republicans argue that media coverage created a greater focus. I would add that the notion that the Democrat's acceptance speech was all Vietnam all the time was fueled by a Republican communication team that was relentlessly on message.

Were there decisive moments that could have turned the election for the Democrat or shoved it out of reach for the Republican? How important were the Kerry gaffes including the words that would live on in Republican ads and speeches "I actually voted for the $87 billion before I voted against it." Or the moment that Kerry decided to windsurf off Nantucket in August within view of photographers who in effect handed the Republican 527s and the Bush campaign a visual metaphor incarnating their claim that Kerry was unmoored from principle, a flip-flopper, out of touch with mainstream values.

Did Bush's petulant performance in the first debate open the possibility that the electorate would rethink its sense of the incumbent? Did the debates benefit Kerry but not enough or too early in the process to affect the final outcome? The Democrats think they did where the Republicans see little net effect. Both teams agree that the debates should be seen as a whole. The Democrats wish they had occurred closer to Election Day.

How did the Republicans close the gender gap? The Republican consultants disclose that they succeeded in finding the 3 points they needed to move from the tie in 2000 to the victory in 2004 from women. As Bush pollster Matthew Dowd put it, "We thought the biggest majority from which we were going to get that 3 points was women, predominantly white women. In the end, if you look at the exit polls, two-thirds of the margin came from white women. We didn't do any better among men on Election Day."

Has the gender gap that closed in 2004 become a marriage gap? The Washington Post's David Von Drehle observes, "In nine states, there are equal numbers of households headed by married and unmarried people. Sure enough, Bush and Kerry split them evenly, four for Bush and five for Kerry—and by middling margins, too: an average 16 points where Bush won, 11 points where Kerry won. Of the 11 states, plus the District of Columbia, where married couples form a minority of all households, Kerry won seven, by a jaw-dropping average of 24 percentage points. Bush won five, by the relatively skimpy average margin of 9 points. The District, with the lowest percentage of married folks, gave Kerry his biggest win: 90 to 9."

Finally, was this an election in which world events such as the 9/11 Commission hearings, the Abu Ghraib scandal, news from Iraq, terrorist attacks around the world, and the last minute Osama tape pushed public perception of the incumbent and his electoral chances more than any imaginable communication by the Kerry campaign could have?

From the debriefing, we learned how the more famous ads from 2004 came to be—from Kerry's on-camera ads, to "Wolves," "SwiftVets," and "Ashley." We heard the debate strategies on each side. We learned what the internal polling numbers said and when they said it, what the strategies of the various consultants were, whether they thought they worked, and how and why the strategies changed. We also learned where they think the press and the scholarly community get the story wrong, from polls based on national samples instead of samples in the battleground states to misunderstandings of the effects of individual debates rather than debates taken as a whole. And we learned that increasingly mass communication exists along side and may be in the process of being displaced by Internet communication and micro-targeting.

Penn Press | Site Use and Privacy Policy
Report Accessibility Issues and Get Help | University of Pennsylvania
Copyright © 2021 University of Pennsylvania Press | All rights reserved