Indiana Law Review The Effects of Media-Based Campaigns on Candidate and Voter Behavior: Implications for Judicial Elections Shanto Iyengar* Introduction In "Television Ads in Judicial Campaigns," 1 Professor Anthony Champagne has made an important contribution to political science. His Paper comprehensively documents the emergence of television advertising as a key ingredient in judicial elections. 2 While traditionally, candidates for judicial positions largely avoided the glare of media attention, 3 more recently would-be judges have begun to conduct campaigns that are strikingly similar to those of candidates for legislative or executive office. 4 Thus, radio and television ads have become the order of the day. 5 Interested observers may well ask what factors underlie the new reliance on this form of campaigning, and, equally important, what consequences will ensue for voters and for the judiciary? This Comment will address both questions from the perspective of a social scientist who has studied the strategies and effects of political advertising in a variety of electoral contexts. My research has focused exclusively on elections for legislative and executive offices. The thoughts offered here are essentially extrapolations drawn from a series of empirically based studies of conventional types of advertising campaigns. I. Why Advertising Now? There are several possible explanations for the increasing importance of television advertising in judicial elections. The first and most basic is that judges, like candidates for any other elective office, have to make their case to voters. Thus, a candidate has to acquire both name recognition and political acceptability. "Free" coverage in the form of news reports is generally * The author holds the Chandler Chair in Communications at Stanford University. Professor Iyengar has written extensively on political campaigns, the effects ofmedia presentations on voter behavior, and the psychology ofpublic opinion. This Paper was prepared specifically for the Symposium on Judicial Campaign Conduct and the First Amendment. The views expressed in this Paper are those of the author and do not necessarily reflect the views or opinions of the National Center for State Courts, the Joyce Foundation, or the Open Society Institute. Supported (in part) by a grant from the Program on Law & Society of the Open Society Institute, as well as a grant from the Joyce Foundation. 1. Anthony Champagne, Television Ads in Judicial Campaigns, 35 IND. L. REV. 669 (2002). 2. Id 3. See id. 4. See id. 5. See id. at 670. 692 INDIANA LAW REVIEW [Vol. 35:691 unavailable tojudicial candidates. 6 Although the press occasionally feels obliged to report on civic affairs, the only beneficiaries of "free" news coverage tend to be candidates contesting "high profile" elections. 7 Virtually from necessity, therefore, judicial candidates must gravitate to the forum of paid advertising. This situation is particularly unfortunate and problematic for challengers, since advertising is perhaps their only available strategy for overcoming the huge incumbency advantage in judicial elections. 8 A second factor that accounts for increasing advertising efforts by judicial candidates is the expansion of the political consulting industry. As elections have become increasingly "professionalized" across the board, the focus of campaign managers has been on the adroit use and manipulation of the media—through "free" coverage, where possible, and through paid advertising in abundance. In fact, a common ploy used by consultants is to use their advertisements as the "bait" with which to attract the attention of reporters. A particularly hard-hitting attack on the opponent is generally worth a news report or two, thus gaining the candidate additional "free" exposure. Due to the strong position of incumbents, challengers injudicial elections have special reasons to seek this type of consultation. Incumbents must then respond in turn with equal media coverage. Thus, it is reasonable to assume thatjudicial elections are only going to become more "sophisticated" in terms of this spiral of advertising and fund raising. II. Voter Behavior in Low Information Elections How might the use of campaign advertising affect the outcome ofjudicial elections? Before I address the possible effects ofadvertising on voter behavior, it is important to acknowledge thatjudicial elections are typical ofwhat political scientists call "low-information" elections, elections for offices about which the public is relatively uninformed. What is especially interesting about these elections is that for the most part voters do make choices.9 Many more, however, "make do" with what little information they have. Therefore, for persons interested injudicial elections, it is paramount to understand how and why voters choose between candidates when they know very little about the "substantive" credentials that would seem most relevant to the candidates themselves. 6. For evidence on the shortage ofnews coverage accorded state and local candidates, see M. Kaplan & N. Hale, TV News Coverage of the 1998 California Gubernatorial Election (unpublished paper, USC). 7. Stephen Ansolabehere, Roy Behr & Shanto Iyengar, The Media Game: American Politics in the TV Age (1 991). 8. Melinda Gann Hall, State Supreme Courts in American Democracy: Probing the Myths ofJudicial Reform, 95 Am. POL. SCI. Rev. 315, 317-18 (2001). 9. See id. at 324. To some degree it is true that voters simply "opt out" by failing to cast a vote on low-information races and propositions; this phenomenon is known in our trade as "ballot roll-off." 2002] A RESPONSE TO PROFESSOR CHAMPAGNE 693 The evidence is overwhelming that in the case of partisan elections, the answer for most voters is simple: voters rely on their party affiliations on the assumption that the candidate of their party is more responsive to their preferences. 10 Contrary to conventional wisdom, party voting is alive and well in the United States. 11 Non-partisan elections provide a greater challenge. When voters are denied information about the party affiliation of judicial candidates, the lack of information could be an overwhelming problem. Yet, this is not so in practice. Social psychologists have demonstrated that human judgment is remarkably resilient and resourceful. 12 In exercisingjudgment, humans tend to reach the best possible outcome given the available resources, under a theory social psychologists refer to as "satisfice." 13 Thus, when denied partisan cues, as in the case of California referenda elections or non-partisan judicial contests, voters likely fall back on relevant, low-cost substitutes, such as endorsements by well- known public figures or voters' beliefs about the groups supporting and opposing the measure or candidate. 14 An example ofthis phenomenon occurred in California where the insurance industry sponsored a series of initiatives to "reform" automobile insurance.' 5 While it is doubtful that the voters had examined the texts of the proposed statutory changes, nevertheless, they managed to reject them decisively. 16 Research demonstrated that the car-driving public perceived the industry's interests as contrary to theirs. 17 All they had to know was who sponsored the proposal in order to sway their vote. 18 In the case ofnon-partisan judicial elections, voters may overcome their lack of information about the candidates' experience or professional/and legal credentials by relying on name recognition or by relying on the word ofcredible public figures who have endorsed particular candidates. In many cases, name recognition may provide incumbent candidates with an edge; in other cases, a challenger with the same name as a well-known athlete or entertainer may spell defeat for a distinguished incumbent. Alternatively, voters may resort to the logic of "performance-based" voting. 19 Reasoning that judges are supposed to reduce the incidence of crime, voters may tend to hold incumbent judges 10. See Larry M. Bartels, Partisanship and Voting Behavior, 7952-/996,44 Am. J. POL.SCI. 35, 36 (2000). 11. See id. 1 2. David Schneider et al., Person Perception ( 1 979). 13. "Satisfying" is often contrasted with "optimizing," which assumes decisionmaking in a context of complete information. 1 4. Arthur Lupia, Shortcuts Versus Encyclopedias: Information and Voting Behavior in California Insurance Reform Elections, 88 AM. Pol. SCI. REV. 63, 64 (1994). 15. Id. 16. Id. 17. See id. at 69-72. 18. Id. 1 9. Morris Fiorina, RetrospectiveVoting in AmericanNational Elections (1981). 694 INDIANA LAW REVIEW [Vol. 35:691 responsible for the level ofcrime. 20 In fact, research indicates that the margin of victory for incumbents is significantly eroded during times of rising crime. 21 Conversely studies show that a decrease in crime rate leads to voter support for the incumbent. 22 I cite the examples ofsponsor credibility, party affiliation, name recognition and perceived performance as voting cues only to make the point that voters choose on a cost effective basis, making do with information that is easily available. All drivers know that insurance companies prefer higher premiums; one only needs to scan the ballot to identify the candidates 9 party affiliation, and the mere act of turning on the television or radio is sufficient to provide information about crime. In the context of these voting cues or shortcuts, we must more closely consider the role ofjudicial campaign advertising. Because I know ofno systematic evidence concerning the effects oftelevision advertising injudicial campaigns per se, I turn to the evidence from national and statewide campaigns for non-judicial offices for clues. III. What Effect Does Advertising Have? There are two broad classes ofeffects that political advertising has on voter attitudes. 23 The first has to do with beliefs about and attitudes toward the candidates. 24 Advertising enables a candidate to convey information, set the political agenda, and ultimately, it is hoped, increase his or her share ofthe vote. The second class of effects is more systemic and relates to the electorate's general feelings about campaigns and the electoral process. 25 In particular, there is evidence to suggest that negative campaigning increases voter cynicism, thus contributing to lower voter turnout. 26 While many do not consider political advertising as a serious form of campaign communication, exposure to advertising nonetheless informs voters and makes them more aware of the candidates. 27 Even when the message is delivered in the form of a thirty-second commercial, embellished with musical jingles and eye-catching visuals, viewers manage to acquire new and relevant 20. See Hall, supra note 8, at 324. 21. See id. at 322. 22. See id. 23. Shanto Iyengar & Adam F. Simon, New Perspectives and Evidence on Political Communication and Campaign Effects, 51 ANN. PSYCHOL. 149, 154 (2000). 24. See id. 25. See id. 26. See id. at 1 52; see also Craig Leonard Brians & Martin P. Wattenberg, Campaign Issue Knowledge and Salience: Comparing Reception from TV Commercials, TV News, and Newspapers, 40 AM. J. POL. SCI. 172, 174 (1996). 27. See David Weaver& Dan Drew, Voter Learning in the 1 990 Off- Year Election: Did the Media Matter?, 70 JOURNALISM Q. 356, 365 (1 993); see also Brians & Wattenberg, supra note 26, at 185. 2002] A RESPONSE TO PROFESSOR CHAMPAGNE 695 information about the sponsoring candidate. 28 In one well-known example, a candidate for U.S. Senate in California managed to recite his position on six different issues in the span ofthirty seconds! 29 One possible explanation for this beneficial effect ofads, it must be acknowledged, is that most viewers have very little prior information about judicial candidates. Given this modest baseline, exposure to campaign advertising cannot help but educate voters. In addition to providing voters with information about the candidates, an important goal of advertisers is to set the campaign agenda. Moreover, most voters are likely to rely on information that is available when it comes time for them to express their preference. Issues or themes that are more frequently encountered during a campaign become more available and salient to voters.30 The reason for repeating the same ad (and Professor Champagne's paper vividly documents the extent of repetitive advertising injudicial elections) 31 is to make the subject matter of the message more prominent in the public agenda. His evidence shows seventy-six percent of the ads aired in judicial races in 2000 referred to criminal or civil justice issues. 32 Given this high percentage, and the fact that advertising was probably the only source of information about the candidates for most voters, most voters likely considered crime rates when casting their vote. Like candidates for other offices, judicial candidates will also utilize advertising to "set" a political agenda. Typically, candidates base such agendas on the issues for which they enjoy a comparative advantage. 33 Thus, by running ads on the subject of crime, candidates accomplish two objectives: first, they make voters think about crime as a relevant issue. Second, they propel voters' beliefs about the two candidates as crime fighters into key determinants ofvote choice. Crime becomes the principal yardstick for evaluating the candidates, thus benefiting the candidate who claims to be "tougher" on crime.34 Alternatively, the candidate who has greater appeal in the area of "family values" or other personal attributes can be expected to attempt to focus the campaign accordingly. By setting the agenda and altering the foundations ofvote choice, campaign advertising can indirectly bolster a candidate's support at the polls. 35 However, one must also address the more immediate question of advertising's effects on 28. See Brians & Wattenberg, supra note 26, at 185. 29. Stephen Ansolabehere & Shanto Iyengar, Going Negative: How Attack Ads Shrink and Polarize the Electorate 59 (1995). 30. Iyengar & Simon, supra note 23, at 157. 3 1 . Champagne, supra note 1 , at 675. 32. He notes that 52. 1 7% ofthe ads were about crime control and 23.8% about civil justice. See id. at 687 tbl.2. 33. See Iyengar & Simon, supra note 23, at 156-58. 34. See, e.g., Joanne M. Miller & Jon A. Krosnick, News Media Impact on the Ingredients ofPresidential Evaluations: Politically Knowledgeable Citizens Are Guided by a TrustedSource, 44 Am. J. Pol. Sci. 295, 308-10 (2000) (reviewing the "priming" phenomenon, weighing issues in accordance with their perceived salience). 35. Ansolabehere & Iyengar, supra note 29, at 82. 696 INDIANA LAW REVIEW [Vol. 35:691 voters' attitudes towards the sponsoring candidate? Does advertising alone make a candidate more appealing or his opponent less appealing? The answer depends on the type of election. In partisan elections, the effects of advertising on candidate preference are contingent on viewers' party affiliation. 36 Ads aired by Democrats are highly persuasive among Democratic voters, less persuasive among non-partisans, and not at all persuasive among Republicans. 37 In the context of partisan judicial elections, accordingly, one would expect that campaign advertising would polarize the electorate by party. In non-partisan elections, where to be recognized is to be liked, advertising should produce a bigger swing in electoral fortunes. In addition to affecting voter choice, campaigns also affect the predispositions underlying the decision to vote. 38 For many years, it was taken for granted that getting out the vote was the sine qua non of effective campaigns. 39 However, campaign managers are well aware that it is often easier to dissuade people from voting than to convert them from the ranks of the opposition. 40 Hence, campaign managers routinely use negative advertising with the specific intent of depressing turnout among the opponents' expected supporters. 41 When attacked, candidates must respond in kind. Thus, the initial attack advertisement spawns a negative campaign, fostering cynical attitudes about the candidates and the political process and lowering turnout. 42 Carefully controlled experimental studies demonstrate that exposure to negative advertising makes voters less likely to believe in the value of elections and more likely to stay home. 43 In summary, advertising enables candidates to become better known, to focus attention on particular issues, and, in many cases, to cast aspersions on the opponent's candidacy. The increase ofnegative advertisements tends to diminish the public's already weak interest in voting. IV. Policy Implications I do not have the space in this Comment to address the many important implications of the modern trend toward media-based judicial elections. However, what I can offer is the prediction that the use of negative campaign tactics injudicial races will spread. Tactics that "work" for consultants in non- judicial campaigns will, inevitably, be put to use in judicial races. Professor 36. Id. at 65. 37. See id. at 64-66. 38. See id. at 104-12. 39. See id. at 146. 40. See id. at 109. 41. Id. 42. Wat 109-10. 43. See generally id; D.A. Houston et a!., Negative Political Advertising and Choice Conflict, 5 J. Experimental Psychol. Applied 3 (1999). 2002] A RESPONSE TO PROFESSOR CHAMPAGNE 697 Champagne's evidence demonstrates that third party advertisers, a major force injudicial elections, have especially strong incentives to "go negative." 44 Since these groups are immune to official oversight or sanction, 45 the "attack-rebuttal- counter-attack" syndrome will characterize increasing numbers ofjudicial races. The spread ofnegative campaigning injudicial races is likely to have adverse consequences for the court system. The motives ofjudicial candidates will be cast into doubt, and public esteem for the judiciary will suffer. Not only will candidates for judicial office be equated with ordinary politicians, but the impartiality, independence, and professionalism of the judiciary will also be called into question. Large-scale advertising in state judicial elections will further politicize state courts in the eyes of the public. V. Possible Remedies If past experience is a guide, the media is not the answer to the problem. Despite a chorus of calls for "free" airtime for candidates, television and radio stations have been reluctant to oblige. There is simply too much money at stake. Attempts to rely on news organizations as referees or arbitrators of judicial campaigns are, unfortunately, likely to prove counterproductive. In recent years, news organizations have taken to running "adwatch" reports in which particular advertisements are subjected to critical scrutiny. 46 One wonders about their efficacy injudicial campaigns. Available evidence indicates that when the news media gets into the fray, the swirl of charges and countercharges is only amplified, and voters become still more cynical and withdrawn. 47 Thus, even if the media were to take up specific issues (e.g., "Did Candidate X really rule in favor of 'drug pushers,' as alleged by Candidate Y?"), it would not solve the problem. Also, "voluntary restraint" will not be an effective method of toning down campaigns because there is a lack of enforcement problem. Candidates may claim to abide by the prescribed code ofconduct, but their surrogates are free to do as they please. In general, candidates are self-interested and rational actors; they pursue winning strategies, not the civic good. Rather than increased media coverage ofjudicial elections or promulgation ofvoluntary canons ofcampaign conduct, the most promising route to campaign reform may be one which would bypass the media entirely by allowing judicial candidates to communicate directly with the electorate. In a few states including California, the secretary ofstate publishes a "judicial guide" which is sent (along with the more comprehensive "voting guide") to every household with a registered voter. The official guide provides background information onjudicial candidates including their educational accomplishments, legal and judicial experience, and professional affiliations. Given the relatively low level of 44. See Champagne, supra note 1, at 673. 45. Id. 46. Ansolabehere & Iyengar, supra note 29, at 1 37. 47. See id. at 140 (discussing statistical evidence available on this point). 698 INDIANA LAW REVIEW [Vol. 35:691 interest in civic affairs, it is likely that very few voters use these guides. By enlivening their content and presentation, technology offers the possibility ofgreater public exposure. A multimedia compact disc, for example, allows voters the opportunity to hear from the candidates "in person" either individually, in the form of prepared statements, or jointly, in the form of debates. The compact disc also provides the user with the freedom to select material that is relevant or interesting. It is both simple and inexpensive to produce a compact disc containing accessible, attention-getting, and relevant information about every judicial election in the state. By presenting the information in a visually appealing and eye-catching manner, electronic voter guides have the potential to broaden the audience for judicial candidates. 48 In states lacking official guides, nonpartisan organizations (such as the League of Women Voters or the state bar association) could be asked to sponsor a similar effort, thus lending credibility to the information. My enthusiasm for a "high tech" approach to voter information is based on more than mere speculation. During the 2000 presidential election, Stanford University, with the full cooperation ofthe Bush and Gore campaigns, produced a multimedia compact disc containing the speeches, televised advertisements and debates, and platforms of the two major candidates. The compact discs were mailed to a representative sample of adult voters two weeks before the election. Nearly one-half of them actually used the compact disc! Subsequent research indicated that voters who used the compact disc were significantly more likely to take an interest in the campaign and vote. 49 As this one example proves, bypassing the media is beneficial to voters and candidates alike. Conclusion If past experience in non-judicial elections is a guide, the use of advertising injudicial campaigns will only increase. Negative advertising is an important ingredient ofadvertising strategy. However, after seeingjudicial candidates and their surrogates hurling charges and countercharges at each other, the public will probably think less of the candidates, the selection process, and the judiciary. How should society respond? Regulating political speech is a non-starter. Depending on the news media and campaign consultants to forego their private interests in favor ofthe public good is unrealistic. The more appropriate remedy is to liberate both candidates and voters from these interests. Modern information technology allows judicial candidates to deliver vast amounts of 48. The compact disc approach has many advantages over the Internet. Despite their profusion, political websites have attracted relatively small audiences. One ofthe primary problems has been that political content on the web is far less appealing than non-political content. In addition, the multimedia content offered by candidate websites requires a level oftechnology (e.g. high speed data transmission capacity) unavailable to most voters. 49. See Shanto Iyengar et al., Does Direct Campaigning Empower Voters?: Estimating the Impact ofa CD Intervention in the 2000 Campaign (2000) (presented at the annual meeting of the American Political Science Ass'n) (on file with author). 2002] A RESPONSE TO PROFESSOR CHAMPAGNE 699 information to a rapidly growing segment of the electorate, free ofeconomic or strategic constraints. For their part, voters are liberated from editorial and other gatekeepers; rather than waiting passively, and most likely in vain, for news reports or advertisements to provide coverage of relevant issues, voters can initiate the queries themselves to obtain information that is personally meaningful. Thus, voter autonomy, the breadth of available information, and candidate control over their message are all realized. In the long run, direct campaigning may contribute to the collective good: increasing the number of people who feel good about the process by which they select judges augurs well for the health of the judiciary.