This analysis addresses two important developments in American political campaigns: the increase in the presence of political consultants and their effect on campaigns. In pursuing this analysis, David Dulio focuses on the central questions of whether consultants are changing the nature of campaigns and whether their presence has displaced the role of political parties. The analysis draws heavily on a survey conducted in 1999 among 505 political consultants by American University staff.

Dulio first traces the rise of the consulting industry and its relationship to the rise of candidate-centered campaigns. He argues that technological developments and dealignment combined to make consultants much more important. Computers led to electronic voter registration files. The widespread availability of computers and software programs made it easier to use those files for analyzing the voting patterns of registrants, pulling samples for polls, analyzing survey results, and selecting specific sets of voters to receive direct mail with specialized messages. These abilities became more important to candidates as the number of independents and swing voters increased. When partisan voting was high and split-ticket voting was low, there was little reason to employ these talents, but as partisan voting declined, the hiring of consultants who were skilled in using the technology to find and access these “movable” voters became very important.

The author then focuses on perhaps the central question that troubles many about consultants: Do they tell candidates what to say they believe in? That is, are polls conducted and then candidates largely advocate what is most popular in the political district, regardless of what the candidate believes in? As in The Candidate movie, do candidates just become vehicles for what consultants find will “work” to win elections? Dulio argues that this does not occur. Instead, consultants help candidates figure out how to articulate what they believe in and present it in a manner that is compatible with most of the electorate while downplaying what is not compatible. Surveys are first used to “find which of the ‘candidate's traits match best with the qualities the voters are seeking’ and the issues on which they are the most compatible” (p. 71). “Consultants ‘help candidates understand how, but not where [in terms of issue positions or ideological stance] to position themselves” (p. 84). “Consultants help candidates define their campaign message through extensive research on the electorate and the candidate's opponent” (p. 82). “It is the consultant's job to put the candidate in the best light possible, but they cannot change a candidate's prior voting history or their publicly stated positions on issues” (p. 74).

I have done polling for candidates for 20 years, and Dulio has this exactly right. Polls are conducted to find out if a candidate has an edge on some issues (which he should stress) and not on others (which he should try to avoid). If most voters are pro-choice and she is pro-choice, and the opponent is pro-life, this position must be stressed. If the public is opposed to tax increases, but the incumbent voted for a tax increase during a recession to provide state aid for schools, then this poll information is valuable for knowing that the incumbent is vulnerable on this issue and that the issue has to be framed as making a tough choice only for the sake of helping the schools. The incumbent cannot change her position, but having knowledge of how this could be a problem means taking the initiative to try to positively define the action. Some issues can be exploited (pro-choice on abortion) and some must be justified to the electorate. A vote for a tax increase cannot be changed or ducked, but it can be explained. The analysis of voting districts and the technologies of direct mail then make it possible to focus relevant messages for specific audiences.

While the argument is well presented by the author, this section could have been improved somewhat with one or two case studies indicating how it plays out in specific situations. That might help convince skeptics that polls do not determine the positions of candidates, and it might demonstrate how polling information helps candidates decide how to present their views and explain past decisions. While that would help, Dulio still does an excellent job presenting his general argument.

The next major issue is the relationship between consultants and political parties. Dulio presents a nuanced and interesting argument. He argues that consultants first became important because parties were struggling with the role they should play as technology and dealignment emerged. Candidates embraced the skills of consultants simply because they needed some way to get to voters. He argues that parties, adapting in general to a changing world, then began to move to incorporate consultants, drawing upon their resources to help their party candidates. The national parties initially tried to develop “in-house” capabilities by building media studios, but they soon realized that keeping up with changing technologies was expensive and that it would be better to contract with private consultants to draw on their skills, while letting the private entities pay the costs of upgrading equipment. Eventually, parties realized that they could engage in a division of labor in which the parties themselves raise money and direct it strategically to contests that are most important to the party and might be won, but allow consultants to do what they do best: use technology to conduct polls, analyze results, and target messages to voters (p. 125). His argument is that parties and consultants, despite the popular image, are not adversaries but that ultimately they have become allies, with each doing what they do best. Again, some examples of how the two interact in specific situations might help, but the argument is clear and persuasive.

In For Better or Worse, Dulio develops an analysis of the role of consultants that focuses on the central issues of whether they are altering campaigns and pushing political parties aside. He produces a well-developed argument that consultants have not taken over campaigns in some sense, but that they serve the goals of candidates and parties. It is an argument worth reading.