Seven _____________________________________________________________

Conclusions: Consequences for Democratic Governance

ELECTIONS ARE THE PRIMARY mechanism by which citizens in a democracy express their wants and desires to their elected officials, and it is through political campaigns that this interaction is managed. Campaigns oblige politicians to define their policy priorities, inform the electorate of the policy alternatives offered by opposing candidates, and provide a forum for policy debate, discussion, and change. We have argued that information about the voters shapes campaign messages and candidate strategies. And information from the campaign influences voter decision making. The dynamics we observe in a presidential campaign, in other words, reflect a reciprocal flow of information and influence between voters and candidates, with this relationship governed by the amount, type, and quality of information each has about the other. This broad perspective of presidential campaigns provides insights into why candidates emphasize some issues instead of others, why some voters are more likely than others to be responsive to those appeals, and ultimately, why we observe the dynamics that we do.

Building on a diverse body of research in political psychology, political communication, voting behavior, and campaign strategy, we have offered three key theoretical propositions in this book:

1. Individual-level responsiveness to presidential campaigns depends on the strength and consistency of voters’ predispositions and on the issue context of the campaign dialogue. Voters facing competing considerations, especially between policy preferences and party identification, are more likely than other voters to rely on campaign information when making up their minds.

2. In an attempt to build a winning coalition between their base supporters and persuadable voters, candidates, motivated by electoral concerns, will target cross-pressured partisans and Independents. The candidates will highlight issues on which these voters disagree with the position taken by the opposing party candidate. In other words, candidates deliberately use wedge issues as part of an electoral strategy.

3. New information and communication technologies have enabled candidates to microtarget different policy messages to different voters, thereby increasing the prevalence and precision of wedge campaign messages.

Throughout this book, we have evaluated our theoretical expectations with a variety of methodological approaches. We have relied on both original and secondary data sources, and have used both quantitative and qualitative methods of analysis. We have covered a lot of ground and presented a number of empirical findings:

1. Partisans are likely to disagree with their preferred political party on policy issues more often than is generally believed. This finding is robust across different policy measures, surveys, and even among the most sophisticated partisans. In addition, contemporary partisans are more likely to hold policy disagreements on cultural, rather than economic, issues.

2. Persuadable voters are not a homogenous group of unsophisticated and indifferent policy moderates, as has often been believed. Rather, persuadable voters hold diverse policy preferences, making it less clear which candidate offers a better match.

3. When exposed to campaign information, persuadable partisans are more likely to be undecided about their presidential vote choice, more likely to change their mind over the course of the campaign, and more likely to defect at the ballot box.

4. Because the campaign helps determine which issue preferences receive greater weight in the vote decision, the content of campaign dialogue shapes who supports which candidate and why. Partisans who disagree with their party on an important policy issue are more likely to defect if that issue is the focus of campaign dialogue.

5. Contrary to expectations that candidates will avoid divisive policy positions or target policy messages to core partisans alone, candidates deliberately attempt to prime wedge issues in order to win over persuadable voters.

6. Candidates are more likely to use divisive wedge issues when they have more information about the preferences of the voters and when they are able to narrowly target their campaign messages.

Taking into account the incentives, interests, and behaviors of both candidates and voters offers a more comprehensive understanding of the role of campaigns in American democracy, and this broader perspective leads to new insights about the dynamics we observe in contemporary campaigns. Perhaps most fundamentally, these arguments have direct implications for how we understand the relationships between citizens and their representatives, as well as the nature of democratic governance more generally. The question of whether political candidates are responsive to citizens’ preferences is at the very heart of how democracy functions, and it is in the context of political campaigns that we can find the answer.

Much of the theoretical and empirical work in political science tends to view the relationship between the governed and governors as dominated either by a “top-down” or “bottom-up” flow of influence. According to the top-down perspective, mass-elite linkages are characterized by elites leading (some would say manipulating) a passive citizenry. In contrast, the bottom-up perspective holds that candidates simply represent the preferences of the public, with the electorate holding accountable any politician who deviates from the public’s preferences. In other words, the basic question is whether political candidates reflect public opinion or shape public opinion. Our findings suggest that the flow of information and influence is neither purely a top-down nor bottom-up process. Rather, campaigns reflect a reciprocal relationship between candidates and voters.

Candidates do not manipulate voters to change their policy attitudes or to vote against their preferences. Rather, campaigns help voters translate their predispositions into their candidate selection by increasing the salience of one consideration over another. To be sure, some might view priming effects as a form of manipulation. Yet responsiveness to campaigns still fundamentally depends on the voters’ own preferences. Further, unresponsive voters are not simply those blindly following their partisan loyalties. Policy-congruent partisans are quick to support their party’s nominee, but our findings suggest that they would also reconsider that support if the party or candidate were to change policy positions. Socially conservative Republicans were among President Bush’s core supporters in 2004, for example, but we would expect them to become persuadable voters if Republicans nominated a pro-choice candidate and Democrats nominated a pro-life one, and the issue of abortion was a salient issue in the campaign. In this respect, then, all voters are potentially persuadable. This conclusion mirrors that of H. Daudt’s classic study of floating voters:

Is it perhaps so that everyone is a potential floater? … Saying that all enfranchised persons are potential floaters is not the same as saying whether persons actually float and, if so, how many of them. This will depend on the political problems, the ways the political parties propose dealing with them and the voters’ reactions to these proposals. Consistent voting behavior on the part of persons or groupings may, then, imply that they are satisfied with the way in which their party approaches the political problems.¹

Our conclusions also differ from those who propose a bottom-up process. According to this perspective, the flow of influence is reversed, with political candidates responsive to voter preferences. Politicians are “like antelope in an open field, they cock their ears and focus their full attention on the slightest sign of danger.”² President Bill Clinton, for instance, has been described as “a weathervane, constantly shifting to and fro in response to the fickle wind of opinion polls.”³ Although this perspective paints a more optimistic picture of electoral accountability, our findings suggest that candidates are not in fact equally responsive to all citizens. Indeed, instead of moving to the center of the ideological spectrum across policy areas, we find that candidates build electoral coalitions by surgically targeting narrow constituencies of interests. From a pluralistic perspective, a candidate attempting to reach out to a diversity of voting blocs should be applauded because of the potential for increased responsiveness.⁴ Yet many of the classic criticisms of pluralism also apply within the electoral context. Just as interest group politics raises concerns about the scope and bias associated with the interests that are represented during policy making, so does a campaign dominated by pluralistic strategies. We consider some of the most troubling potential consequences.

Implications of the Microtargeted Wedge Strategy

The full impact of this changing information environment on candidate and voter behavior is not yet known, but there are at least three potentially grim prospects for the impact of a microtargeted campaign on American democracy: (1) political inequality; (2) superficial politics; and (3) a crisis in governance.

Political Inequality

We have argued that one consequence of increased efficiency in campaign targeting is that candidates are less likely to mobilize individuals who are unlikely to vote. This has clear implications for government policy, as a long and rich literature about the participation gaps in American politics documents. All voters are not equal in the eyes of presidential candidates.

Candidates are particularly responsive to those voters who are strategically valuable in building a successful, short-term electoral coalition. Candidates are simultaneously (and intentionally) ignoring the preferences of those who are not critical to their coalition’s electoral success. As candidates increasingly rely on sophisticated technology to narrowcast their campaign messages to strategically chosen voters living in strategically chosen states, the issues of interest to these voters outweigh all others.

When campaigns attempt to reach out to the full electorate, a candidate must harmonize and synthesize interests, incorporating them into a policy message that resonates with the general interest of the nation. The extent to which contemporary microtargeted campaigns give priority to the needs and desires of only the electoral expedient citizens affects the degree to which the electoral process fails to live up to the democratic ideal. As Steven Schier has argued, “The narrow strategic focus of activation makes majority rule at best an incidental byproduct of this system. Candidates seek to win elections by targeting a small group of swing voters in search of a plurality of those who vote, not a majority of all citizens. Groups have little incentive to command majority opinion if they can prevail without it—and they often can.”⁵ These microtargeted campaign strategies exacerbate inequalities in the American political system, contributing to a system of democracy “of and by” a myriad of narrowly focused groups of swing voters, not democracy “of and by” the people.

Superficial Politics?

Indeed, even when candidates are attentive to a particular group of voters, candidates will emphasize wedge issues because they help create a strategic advantage, not because they are necessarily the most important issues to the targeted voters. Such voters are not so much being manipulated as induced (some might say bribed) with promises to enact a preferred policy. What might be a rational behavior on the part of the individual voter is perhaps not the best policy for the collective. In 1996, for instance, Bill Clinton was often derided for emphasizing “bite-sized, or ‘small-bore’ initiatives … too puny for presidential action.”⁶ With detailed information about individual voters, candidates are able to activate narrow policy interests that might be considered superficial, or at least less significant, than those issues of concern to the broader public.

The 2004 presidential election highlights the potential disconnect between the priorities of the public and campaign messages. According to the 2004 National Annenberg Election study, the typical respondent consistently reported that the issues of unemployment, the War in Iraq, terrorism, and health care were at the top of their “most important problem” list.⁷ Among Democrats, Republicans, and Independents, these same four issues were consistently mentioned as the policy areas that citizens believed were the most important. In fact, only 3 percent of Democrats and 5 percent of Independents reported that a “lack of moral/family values” was among the most important issues during the 2004 presidential elections. Yet, we found that 19 percent of direct-mail issue appeals targeted to Democratic voters focused on moral issues, and 24 percent of direct-mail issue appeals targeted to Independents emphasized moral wedge issues. Clearly, the policies emphasized in the 2004 ground-war campaign did not reflect the issue priorities that the general American public believed to be most important. Instead, the targeted messages emphasized wedge issues that candidates believed offered them a unique advantage relative to their opponent. To the extent that candidates emphasize these narrower policy interests during their campaign, the energy and resources remaining that may be devoted to other policy areas are limited. And once elected, the winning candidate has the difficult task of turning fragmented campaign promises into governing priorities.

A Crisis in Governance?

Following the 2004 election, President Bush proclaimed that “the people made it clear what they wanted. … Let me put it to you this way: I earned capital in the campaign, political capital, and now I intend to spend it.”⁸ Such political rhetoric notwithstanding, if different individuals intend their vote to send different policy messages, interpreting election outcomes as an indicator of what “the public wants” misses the fact that election outcomes represent a cacophony of divergent policy goals. Quite simply, interpreting the meaning of an election is difficult when there is a fragmented and narrowly tailored campaign dialogue (not to say anything of narrow electoral margins).

Popular elections are often believed to provide presidents with the political leverage to forge policy changes. Under the model of responsible party government, political parties should present distinct policy alternatives, so that a vote cast for one candidate over the other provides a clear signal of the voters’ preferred policy direction. In 1952, for instance, Republican General Dwight D. Eisenhower campaigned on one key promise—to end the war in Korea, giving him an unambiguous popular mandate when he won. When candidates present a clear and limited set of policy alternatives, communicated to the entire electorate, the winning candidate has a greater claim that his policy priorities reflect the will of the people.⁹ Yet in contemporary presidential campaigns, a fragmented and diverse policy agenda undermines the potential for an election outcome to indicate public support for any one policy.

It is difficult to construct a sustainable notion of electoral accountability without a shared public discourse on the candidates’ policy positions and future agendas. Democratic accountability rests on some minimal conditions that voters at least know their own policy preferences and the policy positions of candidates. Although political scientists have long argued that individual citizens do not always live up to this standard, scholars have often taken comfort in the finding that politicians are responsive to collective public opinion, which is thought to be more informed, stable, and rational.¹⁰ At its extreme, a fragmented campaign dialogue undercuts the very notion that a collective public opinion even exists. Peter Swire, the Clinton administration’s chief counselor for privacy, explained,

In the nightmare, every voter will get a tailored message based on detailed information about the voter. … [This] means that the public debates lack content and the real election happens in the privacy of these mailings. The candidate knows everything about the voter, but the media and the public know nothing about what the candidate really believes. It is, in effect, a nearly perfect perversion of the political process.¹¹

How do we maintain electoral accountability if democratic deliberation occurs alone rather than as a public collective?

In 2004 journalists and pundits concluded that the public supported Bush because of his conservative stance on moral issues like gay marriage and abortion. The London Times, for instance, reported that “Americans voted in record numbers for a Republican president primarily because they identified with his moral agenda.”¹² In fact, what “the people” said they wanted was not very clear at all because campaign dialogue was reflective of individual, rather than aggregate, preferences. Empirical analyses of voter behavior in the 2004 election have found that the public did not offer Bush a “moral mandate”—the average voter cast her ballot on the basis of economic considerations, party identification, the War in Iraq, and concerns over terrorism.¹³ Of course, the story of the average voter does not fully capture the more complex and nuanced relationship between voters and candidates in the 2004 election.

As the broader campaign was being waged primarily on issues of the economy and war, millions of Christian social conservatives were told that moral issues were at stake in the election. The Republican National Committee, for instance, mailed voters in Arkansas and West Virginia flyers that printed the word allowed over a picture of a same-sex couple and banned over an image of the Bible. Simultaneously, the Bush campaign targeted elderly investors, military families, small-business owners, and so on. For the state of Michigan alone, the Republican Party created a 157-page report separating likely voters in this battleground state into dozens of separate “microtargeting segments”—ranging from “tax-and-terrorism moderates” to “traditional-marriage Democrats” to “terrorism and health care Democrats.” Once voters were identified in these specific groups, they were targeted with messages that directly corresponded to the issues that predisposed them to support President Bush.¹⁴ Different segments of the population were each told that some issue they cared about was a top priority of President Bush. This type of segmentation means that any interpretation of what the election was “about” was incomplete because there was a multiplicity of policy agendas presented to the public and a multiplicity of different agendas supported by voters.

When an election is finally over and the ballots are counted, the winning candidate must decide which policies to pursue. Yet, candidates’ campaign agendas are largely divorced from strategic decisions regarding how candidates might successfully govern if they are elected. Communications scholar Nicholas O’Shaughnessy argues that “coalitions of support are created with much greater ease than in former times, for the thrust of the new technology is towards segmentation and its results. … [G]roups can now be solicited as individuals on their key interests and enthusiasms; but the loyalties of such coalitions are also more fickle since they no longer depend on organic linkages to political parties.”¹⁵ Inevitably, each of the targeted segments expect the candidate to fulfill the promises made to them during the campaign. Following the 2004 campaign, Rev. D. James Kennedy, a broadcast evangelist, declared, “[N]ow that values voters have delivered for George Bush, he must deliver for voters. The defense of innocent unborn human life, the protection of marriage, and the nomination and confirmation of judges who will interpret the Constitution, not make law from the bench, must be first priorities come January.”¹⁶

In many ways, the microrecruiting of various voting coalitions raises problems similar to the concerns raised by John Mueller and his “coalition of minorities” theory of presidential leadership.¹⁷ Mueller argued that it was inevitable for presidential popularity to decline over the course of an administration, as any presidential decision would antagonize at least some group of supporters. Voters supporting different policy interests might come together temporarily for electoral purposes, but their solidarity is severely tested when it comes to policy making. Once the pressure of governing becomes real, these loose electoral coalitions are likely to break, often leaving the governing party without substantial leverage to accomplish its goals. While Mueller has noted that the negative trend in presidential popularity affects nearly all presidents, contemporary campaign strategies may exacerbate this pattern.

The Terri Schiavo incident, for example, clearly laid bare the governance fractures within the Republican Party’s 2004 electoral coalition. Terri Schiavo, a Florida woman in a persistent vegetative state for fifteen years, ended up at the center of a legal battle over euthanasia. Schiavo’s husband battled with her parents over the right to make life-terminating decisions, and the courts ultimately sided with him. In March 2005 the Republican majority in Congress, with President Bush’s approval, returned from legislative recess to pass a special bill—after midnight on a Sunday night—to prevent doctors from removing Schiavo’s feeding tube. The courts ultimately rejected the bill, but the move angered many of Bush’s supporters. A Wall Street Journal/NBC News poll found that 48 percent of Republicans surveyed thought that reinserting Schiavo’s tube was the “wrong thing to do” compared to 39 percent who said it was the “right thing to do.”

Conservative commentator and former Republican Congressman Robert Barr criticized the federal government intervention, “To simply say that the ‘culture of life,’ or whatever you call it means that we don’t have to pay attention to the principles of federalism or separation of powers is certainly not a conservative viewpoint.”¹⁸ While the issue of euthanasia was an important concern to some Christian conservatives, others in the Republican electoral coalition were frustrated to see government attention distracted from problems like the War in Iraq, terrorism, tax reform, or deficit reduction. Also revealing is the timing of the Schiavo incident, which came just after a failed attempt to reform social security—a campaign promise to investment-minded Republicans that also found mixed support among the larger Republican coalition.

The political fallout that President Bush and congressional Republicans endured so quickly after claiming mandates from the voting public shows just how difficult it can be to maintain electoral coalitions once governing begins.¹⁹ Although Christian evangelicals were supportive of Bush and the Republican Party during the Schiavo hearings, they would later become frustrated by Bush’s lack of action on moral issues as his attention turned to other priorities. According to one report, “In the last several weeks, Dr. James C. Dobson, founder of Focus on the Family and one of the most influential Christian conservatives, has publicly accused Republican leaders of betraying the social conservatives who helped elect them in 2004. He has also warned in private meetings with about a dozen of the top Republicans in Washington that he may turn critic this fall unless the party delivers on conservative goals.”²⁰

Even with a GOP-controlled Congress, President Bush found it difficult to fulfill his myriad campaign promises. The 109th Congress was often described as one of the least productive in decades, resurrecting Harry Truman’s taunt of a “do-nothing” Congress. The list of campaign promises failing or receiving scant attention during this administration—ranging from a constitutional amendment to ban gay marriage to social security reform—is substantial. But given Bush’s campaign strategy, perhaps such dissatisfaction should not be surprising. Thomas Mann and Norman Ornstein have lamented, “Campaigning has become increasingly antithetical to governing. Candidates decreasingly use campaigns to build public support for governing decisions or to forge public consensus before making policy. … Candidates often frame campaign themes and take positions in ways that frustrate rather than facilitate the task of governing after the election.”²¹ The strategic decisions that help candidates create winning electoral coalitions do not always translate into successful governing coalitions.

We have painted a bleak future and potentially grim pitfalls for presidential campaigns in a hyperinformation environment. We emphasize these negative consequences in part because so many scholars and political observers have tended to highlight the positive benefits of information technology for democracy. Scholars have argued that information and communication technologies can be used to improve democratic deliberation by creating online forums, by enhancing citizen engagement, and by overcoming political and social inequalities.²² To be sure, a few scholars have offered a more pessimistic view, suggesting that the contemporary information environment has exacerbated gaps in political interest, political knowledge, and political participation. Our analysis shows that technology also carries potential negative consequences for democracy by virtue of the impact on political candidates—particularly in the role technology plays in gathering and communicating information. Of course, these troublesome predictions reach beyond the bounds of our empirical evidence, so we leave it to the reader to evaluate the extent to which technology represents promise or cause for concern.

Beyond U.S. Presidential Campaigns?

Throughout the book we have focused on presidential elections, making it unclear whether the relationships between information, candidate strategies, and voter decision making generalize to lower-level races, such as Senate or gubernatorial elections, or if our arguments extend to the electoral systems in other countries. It is possible that American presidential campaigns offer conditions that might be uniquely suited to wedge strategies. The two-party system means that the parties will always be heterogeneous coalitions, and the prominence of the contest means that voters are more likely to be informed about the policy positions of the candidates and that cross-pressured voters are still likely to show up on Election Day.

In lower-level races, a sufficiently homogeneous district, or perhaps an election in which turnout is dominated by core supporters, candidates can and do avoid a wedge-issue strategy. Indeed, candidates in uncompetitive races may be able to largely avoid controversial issues altogether, instead focusing on personal appeals and name recognition. In these electoral contexts, we might also be more likely to find that potential wedge issues are neutralized by the position taking of the selected candidates. In 2006, for instance, in some congressional districts in which abortion had the potential to be used as a wedge issue by Republican candidates, the Democrats put forward a pro-life candidate—which effectively neutralized abortion as a potential wedge issue. In lower-level races, candidates and parties are at greater liberty to “take off the table” those issues that have the greatest potential for dividing their coalition of supporters. This would be more difficult in a presidential election in which the constituency is more diverse and there are stronger pressures to adhere to the party platform.

Even still, there are a number of lower-level races in which we have observed wedge-issue campaigns. Competitive congressional races, for instance, are often characterized by explicit attempts to appeal to persuadable voters on the basis of wedge issues. In her run against Republican incumbent Senator Jim Talent, Democratic candidate Claire McCaskill emphasized her support for stem cell research in an attempt to reach out to Republican voters who did not share President Bush’s opposition to such research. McCaskill’s message received considerable assistance once a ballot measure about embryonic stem cell research became a focal point of the campaign. According to one journalist account of the campaign, Democrats emphasized stem cell research

to exploit a division between conservatives who oppose the science and other Republicans more open to it. … With the Talent-McCaskill race too close to call, the initiative has thrust Mr. Talent into a treacherous Republican crosscurrent. On one side are Christian conservatives. … On the other are business-minded Republicans … saying the science holds promise not only for patients, but also for the economic health of the state.²³

Eventually, McCaskill, and the ballot measure, received the support of a slim majority of Missouri voters.

As the Senate race in Missouri suggests, the recent growth in the use of state ballot initiatives illustrates an attempt on the part of other political actors besides candidates to prime divisive issues. A recent study reiterated our warning that “wedge initiatives” have potential negative consequences for democratic governance because they are typically motivated by campaign politics rather than policy making. According to Thad Kousser and Mathew McCubbins, “Political parties, individuals and interest groups are all scouring the policy space to find niche issues for the purpose of affecting elections or rewarding or punishing candidates and parties.”²⁴ According to their report, in the 2004 election alone, there were 162 ballot initiatives in thirty-four states. These initiatives highlight the deepest cleavages in the party coalitions. Republican groups have used gay marriage and immigration issues to reach cross-pressured Democrats, while Democratic groups have organized initiatives based on minimum wage and environmental protection in an effort to reach cross-pressured Republicans. These ballot initiatives are often organized and funded by out-of-state groups interested in using controversial issues to win both national and local electoral office. They are anything but the idealized picture of a grass-roots effort of concerned citizens trying to influence public policy.

Similar to lower-level races, it is also unclear if the incentives for using wedge issues remain in the comparative context. Certainly, campaign strategies will reflect the institutional features of a political system, including the basic institutional structure of government, electoral laws, procedures governing elections, and nature of political actors. We might expect, then, that campaigns in proportional electoral systems, by lowering the hurdles to office, would be see fewer divisive issues used in the campaign. The coalition building that we observe within the campaign process in the United States might instead occur after elections among party leaders and elected representatives in other political systems.

On the other hand, recent research has highlighted that many countries have “Americanized” their electoral campaigns. Campaigns are increasingly characterized by everything from an expanded use of consultants and negative advertising to targeted campaign appeals. Evaluations of political marketing in Western democracies, and in the British Labor party in particular, suggest that microtargeting of wedge issues may be on the rise. According to Darren Lillenker,

Over the past four decades, parties across Europe have brought in consultants to support their campaigning, conducted research to aid the design of communication and used techniques associated with branding when constructing symbolic representations of their party. Leaders’ style and image, key messages, party motifs, as well as the various modes of advertising, are all part of sophisticated marketing strategies more associated with producers of fast moving consumer goods than with political parties.²⁵

Obviously, the extent to which the trends we have identified in American presidential elections occur in other electoral and institutional contexts is an important topic for future research. Somewhat more apparent in other polished systems is the more fundamental relationship between the information environment and campaign strategies.

Australia serves as a prime example. It maintains a proportional electoral system requiring compulsory voting of all eligible citizens. Despite these institutional differences, a cursory examination of recent campaign strategies reveals a similar relationship between the information available about individual voters and the use of microtargeted wedge campaign strategies as we have identified in our earlier analysis.

Stemming from the legal requirements for mandatory voter turnout, the Australian Election Commission (AEC) possesses an incredible amount of information about the voting public. The AEC maintains and regularly updates a centralized database, including a registrant’s full name and address, telephone number, date of birth, sex, occupation (optional), mobile number (optional), and date of naturalization. With information from the central registration system as the base, the major political parties in Australia maintain internal databases that match this registration data with other important facts about the voters, including information gathered following contact with politicians, letters to the editor, party donation history, and so on.²⁶ The Coalition’s Feedback database, for instance, includes fields for three hundred different issue interests and preferences.²⁷ The local-party office staff is trained to log all constituency correspondence into the database, although members in safe seats or in their last term appear less motivated to make comprehensive entries. The parties use these data to attract potential supporters; for instance, the local representative will send a welcome letter to individuals moving into the area. The information is also used to tailor campaign messages.²⁸

Using this information, these targeted campaign messages can contain divisive content. The 2001 federal election, for instance, is often cited as a classic example of dog-whistle politics. The Australian economy was in crisis in 2001—several corporations had collapsed, unemployment had dramatically increased, and the Australian dollar plunged to all-time lows—but the campaign communication of the Labor Party focused on issues of border security, immigration, and terrorism. Labor Party advertisements showed a picture of a clenched-fisted John Howard proclaiming, “We decide who comes into this country.” As one political commentator wrote, “Howard’s brand of wedge politics very much follows the tactics used by the US Republicans. … ‘Politically correct,’ ‘free speech,’ ‘family values’ and ‘values neutral’ are key phrases that have made their way across the Pacific.”²⁹

While certainly not definitive, the Australian case is suggestive of the link between the information that candidates have about persuadable voters and their willingness to take a position on a divisive issue in the political campaign.

Voter Response to Microtargeted Messages

Another critical topic for future research is to isolate and understand more completely the influence of wedge campaign messages on voters in the context of a complex campaign environment. What is the impact of microtargeted campaign messages above and beyond the dozens of other mailers, phone calls, and personal visits that bombard residents of battleground states? We have argued that microtargeting is consequential even if it has little impact on the voters because it shapes the specific policy promises candidates make and determines a candidate’s perceived constituency. Yet the prominence of this strategy will undoubtedly depend on its influence. Is the targeted mail piece more persuasive because it is individualized on an issue the recipient cares about? Or is a mailing more easily discarded as a piece of junk mail?

We might actually expect that microtargeted messages are in fact more persuasive than other campaign communications. An individualized and personalized message should be more compelling than the broad-based appeal in a television ad. Further, compared to television advertising, targeted communications receive less scrutiny from the media and opposition camps so they are not only more likely to contain divisive issues, they should also be more likely to contain sensational and inflammatory content. Mike Russell, a spokesman for Swift Vets and POWs for Truth, explained: “If direct mail were food, it would be hot salsa.”³⁰ In 2004, for instance, the Bush campaign was exceptionally careful in campaign commercials to avoid the appearance that they were exploiting 9/11 for political gain. These concerns, however, did not carry over into the messages communicated through direct mail—one mailing in particular contained several images of the destroyed World Trade Center with the caption “How can John Kerry lead America in a time of War?” Of course, Republicans were not the only ones willing to use direct mail to send provocative campaign messages. From the other side of the aisle, Democrats also sent campaign messages and images through direct mail they would not likely want to convey to wider audiences. For example, one Democratic mailing contained an image of a 1960s white fire fighter blasting an African American with a fire hose, with the caption “This is what they used to do to keep us from voting. Don’t let them do it again.”

The extent to which direct mail is a persuasive form of campaign communication is an especially important question for future research because it will help us predict the broader impact of these campaign tactics on election outcomes.

Final Words: Speculations about Future Campaign Technology

We cannot overstate the economic and social impact of new information and communication technologies. A recent study concluded, “As canals and railroads were the infrastructure for the emerging industrial economy in the 19th century, information might be thought of as the infrastructure for the new ‘knowledge economy.’”³¹ And we expect changes in the information environment also have important implications for electoral politics—a political campaign takes place within a specific social and technological context, and this environment shapes the dynamics of the campaign. At different stages in history, the introduction of radio, TV, and other technologies has had a profound influence on the relationship between citizens, government, and the campaigns. Candidates continually look for strategies to gain the slightest edge over their competition, and technological innovations are one way that candidates look for an electoral advantage. It is, of course, difficult to predict exactly what or how new technologies will be used. Regardless, one prediction seems clear—as technological advances continue to change the political landscape at an exponential pace, we expect to see candidates attempting to find and use even more detailed information about individual voters. A recent white paper from one microtargeting firm proposed mining online resumes (there are more than 50 million online) to collect information about where and what someone studied in college. With the emergence of online communities through MySpace, Facebook, and other Web sites, candidates may be able to penetrate social networks to quickly and efficiently target like-minded individuals through trusted personal connections rather than broad, impersonal appeals. Given the link between driver’s-license databases and voter registration lists, it may not be too much of a stretch to imagine that voter databases could eventually include photographs of individual voters. Candidates could then send extremely personalized campaign messages that contain not only the issues an individual cares about but also the individual’s own image. Already, we see direct mailings that are personalized by the race of the recipient—with African Americans receiving direct mail showcasing African Americans, and Hispanics receiving mailings with images of Hispanics. In Steven Spielberg’s science-fiction blockbuster Minority Report, the future—the year 2054—is characterized by individualized advertising based on “Big Brother”–like surveillance. In the political world, that imagined future is perhaps closer than many realize.

At the same time, there is the potential for new technological innovations to limit the effectiveness of microtargeting. New technologies may make it more difficult for candidates to get away with sending blatantly conflicting messages to different voters. It seems possible that as the public (and other candidates) become increasingly aware of the discrepancies in campaign messages between different communication modes, the incentive to use targeted messages could be reduced. With video-sharing Web sites like YouTube, microtargeted stump speeches can be uploaded and shared with a broader audience, bringing the “under-the-radar” messages of the grand war into the public sphere. In the 2006 Virginia election, for example, Senator George Allen was videotaped calling an Indian American a “macaca.” Although Senator Allen apologized, hundreds of thousands of viewers watched the clip on YouTube, and Senator Allen’s lead in the polls began a steady decline and he was ultimately defeated. Of course, accountability via the Internet assumes that audiences are interested in what candidates are saying to different groups and that candidates can be linked to specific messages. It is already common practice for divisive campaign messages to originate with the party rather than the candidate, making it easier for candidates to distance themselves from anything controversial. But the Internet allows for an unprecedented level of anonymous, unregulated political mudslinging and rumor spreading. Early in the 2008 Democratic presidential primary season, for instance, an unauthorized amateur “1984” video attacking Hillary Clinton as Orwellian and promoting Barack Obama was anonymously posted on YouTube. It received nearly 4 million hits in five months. Revealing his identity on the Huffington Post blog two weeks later, the ad’s author observed, “This ad was not the first citizen ad, and it will not be the last. The game has changed.”³²

Whatever the future holds for new information technologies, it seems apparent that the electorate of tomorrow will face an increasingly fragmented and diverse information environment. With hypermedia campaigns, it is ever more difficult for voters to “tame the information tide.”³³ This is especially the case for Independents and cross-pressured partisans who are targeted from both sides of the partisan aisle. Targeted communications contribute to a swirl, some might say tornado, of information in contemporary political campaigns. In this context, the full landscape of policy differences between the candidates can be obscured and it becomes more difficult for voters to remove the “chaff” from the “wheat.”

In this age of narrowcasted campaigns and segmented issue messages, it perhaps becomes increasingly important for the mainstream media to monitor the exchange of information between citizens and their representatives. So fundamental is the media’s role in ensuring democratic accountability—by informing the electorate of the policy debates, distilling complicated events and policies into comprehensible narratives, and reporting on the behaviors and misbehaviors of political leaders—that Thomas Jefferson once remarked, “Were it left to me to decide whether we should have government without newspapers, or newspapers without a government, I should not hesitate a moment to prefer the latter.”³⁴ The media often obsess over candidates’ spending on television commercials, setting up elaborate “campaign ad watches” to ensure the veracity of television campaign commercials, but our analysis makes clear that these communication venues provide only a limited perspective of the candidates’ campaign agendas. The targeted ground-war communication may be more difficult to track, but is more divisive and less connected to the broader issue dialogue of the campaign. The media have an obligation to bring candidates’ rhetoric and promises to the broader public discussion and to expose any inconsistencies between candidates’ broadcast policy goals and their narrowcast promises on wedge issues. While the news media often scrutinize the candidates’ campaign tactics and motivations, they must also consider the overlap between campaign tactics and the policy substance of the campaign.

Presidential campaigns remain the fundamental link between citizens and their government in an electoral democracy. Understanding how voters make up their minds in an election and why candidates offer the policy alternatives they do is critical if we are to successfully evaluate the state of American democracy. We hope to have offered a slice of insight by identifying the incentives behind candidates’ campaign promises and policy position taking and by offering a general perspective of who in the electorate responds to information presented during the campaign. Although American democracy is far from flawless, and new information technologies certainly have the potential to exacerbate the faults, we nonetheless conclude with the observation that the balance of power in American democracy is still held by its citizens. And our analysis suggests that these citizens have the capacity and motivation to deliberate about their vote decision. Although the fragmentation of campaign dialogue has potentially negative implications for political inequality and governance, we remain reassured that candidates are ultimately constrained by voters, even if not by all voters. It is still in the interaction of citizens and government during an American presidential campaign that we find the basic structure, however imperfect, of a democratic process.

---

¹ Daudt, Floating Voters and the Floating Vote, 160–61.

² Stimson et al., “Dynamic Representation,” 559.

³ Berinsky, Silent Voices, 1.

⁴ Robert Dahl, Who Governs?

⁵ Schier, By Invitation Only, 36.

⁶ Sosnik et al., Applebee’s America, 23.

⁷ National Annenberg Election Survey, press release 28 October 2004, “Terrorism Seen as Most Important Problem for Bush Backers; Kerry’s Worry Most about Economy, Annenberg Data Show.”

⁸ Richard Stevenson, “Confident Bush Outlines Ambitious Plan for 2nd Term,” New York Times, 5 November 2004, A1.

⁹ There is a long-standing debate in political science about presidential mandates. See Conley, Presidential Mandates; Grossback et al., Mandate Politics.

¹⁰ Page and Shapiro, The Rational Public.

¹¹ Jon Gertner, “The Very, Very Personal Is the Political,” New York Times Magazine, 15 February 2004, 43.

¹² James Harding, “Electorate Puts Moral Concerns Ahead of Policy,” Financial Times (London), 5 November 2004, A8.

¹³ Hillygus and Shields, “Moral Issues and Voter Decision Making in the 2004 Presidential Election.”

¹⁴ Sosnik et al., Applebee’s America, 34–35.

¹⁵ O’Shaughnessy, The Phenomenon of Political Marketing, 13.

¹⁶ Larry Eichel, “ ‘Values Voters’ Seek Their Reward in Policy,” Philadelphia Inquirer, 7 November 2004.

¹⁷ Mueller, “Presidential Popularity from Truman to Johnson.”

¹⁸ “Political Fallout over Schiavo,” CBS News online, 23 March 2005.

¹⁹ Shailagh Murray and Mike Allen, “Schiavo Tests Priorities of GOP,” Washington Post, 26 March 2005, A1.

²⁰ David Kirkpatrick, “Conservative Christians Warn Republicans Against Inaction,” New York Times, 15 May 2006, A1.

²¹ Ornstein and Mann, The Permanent Campaign and Its Future, 225.

²² Dahl, Democracy and Its Critics; Barber, Strong Democracy.

²³ Sheryl Gay Stolberg, “Democrats Hope to Divide G.O.P. over Stem Cells,” New York Times, 24 April 2006.

²⁴ Kousser and McCubbins, “Social Choice, Crypto-Initiatives, and Policymaking by Direct Democracy.”

²⁵ Lillenker, “The Impact of Political Marketing on Internal Party Democracy,” 570.

²⁶ The ALP’s database is called Electrac; the Coalition’s is called Feedback. These systems date to the 1990s, but did not become fully operational until 1996.

²⁷ Privacy laws prevent community and public organizations from handing out lists of members.

²⁸ Onselen and Errington, “Electoral Databases.”

²⁹ Iain Lygo, “Racism Rather than Relief in Australia,” Z Magazine, 25 May 2004.

³⁰ Glen Justice, “In Final Days, Attacks Are in the Mail and Below the Radar,” New York Times Magazine, 31 October 2004, 30.

³¹ Hillygus et al., The Hard Count, 76.

³² Phil de Vellis, “I Made the Vote Different ad,“ Huffington Post Blog, posted 21 March 2007.

³³ Graber, Processing the News.

³⁴ Kurland and Lerner, The Founders’ Constitution.