Scandals can have serious ramifications for the elected and those striving to be elected. Being associated with a scandal clearly harms the popularity of most politicians, as several studies chart the dip in public approval of a politician following a scandal (Andolina and Wilcox Reference Andolina, Wilcox, Rozell and Wilcox2000; Newman and Forcehimes Reference Newman and Forcehimes2010; Simon and Ostrom Reference Simon and Ostrom1989). Scandals also harm politicians seeking reelection by prematurely ejecting them from the political process or redacting vote share (Basinger Reference Basinger2013; Praino and Moscardelli Reference Praino, Stockemer and Moscardelli2013). For presidents and governors, scandals truncate the time in office served by these elected officials (Rottinghaus Reference Rottinghaus2014). American presidents, in particular, tend to be highly susceptible to scandals as their political fortunes are often linked to such events (Woodward Reference Woodward1999). This is especially true of certain presidents who tend to be shrouded in accusations (Stewart Reference Stewart1996). The resulting legal skirmishes have given an avenue to the venom in partisan politics for a generation since Watergate (Ginsberg and Shefter Reference Ginsberg and Shefter1999). Other work positions the politics of scandal at the center of modern American political life (Abshire Reference Abshire2005; Fousek and Wasserman Reference Fousek and Wasserman2010; Roberts and Doss Reference Roberts and Doss1997).
Yet, despite the apparent impact of scandals on elected officials and congressional candidates, we know little about how the arc of scandal affects the resources or standing of presidential candidates seeking the nomination of their party. For every Gary Hart and Herman Cain who had their candidacies cut short after public scandals emerged, there is a Bill Clinton or John McCain who survive a scandal to go on to win the nomination. Filling this empirical gap is important because the nomination and election process for presidential candidates is a rapid-fire process where daily events may have a significant effect on the success of a candidate. Scandals affecting candidates’ campaign are often exaggerated by the media (Entman Reference Entman2012; Sabato Reference Sabato1993) and are used as a basis for priming the narrative about that candidate (Brody and Shapiro Reference Brody and Shapiro1989). If these events are consequential to the robustness of a campaign or the survival of a candidate, it is critical to understand the precise impacts of scandals in nomination politics. The presumption is that scandals harm candidates for the nomination of their party for president but what is the depth of the damage done?
The purpose of this article is to determine the effect of scandals on the standing and resources of presidential campaigns, all of which are keys to predicting which candidate gets the nomination of their party (Mayer Reference Mayer2000). In short, what effect do scandals have on specific indicators of presidential campaign success? Using individual scandals connected to nonincumbent presidential aspirants from 1996 to 2012, we track the effect of each scandal on the campaigns’ daily fundraising totals, competitive standing in the race, and the number of party endorsements. Using daily data for each of these indicators of campaign nomination success, this article maps a clear and precise effect of scandals on the life of a presidential candidate. Given how truncated the nomination process has become after the first primary is held (Hagen and Mayer Reference Hagen, Mayer and Mayer2000), how quickly negative stories ripple through the media during the nomination process (Haynes and Pitts Reference Haynes and Pitts2009), and the rapidity with which potential donors and the party elite may abandon a nominee, the effect of a scandal on a nominee can have a significant damaging effect on a candidate’s chances.
LOCATING CAMPAIGN SCANDALS
The definition of scandals used in this article involves verified allegations of illegal, unethical, or immoral wrongdoing. The definition includes adultery because of the unique place of inappropriate sexual relationships in the panacea of political scandals (Rosen Reference Rosen2009), especially those on the campaign trail. Including sex scandals ensures that we include some “transgressions” of conventional morality, but we wish to exclude gossip, innuendo, and unsubstantiated rumors of private behavior. This definition is consistent with the sense that scandals involve a “transgression” (Thompson Reference Thompson2000) or an act contrary to conventional morality or “a set of shared values” (Garment Reference Garment1991). For scandals of nonsexual nature, an alleged violation of law or a code of ethics is the threshold criteria. This definition intentionally excludes strategic mistakes, incompetence, unpopular policy decisions, or negative press. This also prominently excludes instances that generate negative publicity, such as missteps when governor Rick Perry in a televised debate could not remember the third executive department he would eliminate (the “oops” moment) (Ceaser, Busch and Pitney Reference Ceaser, Busch and Pitney2013) or gaffes when Mitt Romney bet another candidate $10,000 to correct a point of contention in a debate (Zeleny and Parker Reference Zeleny and Parker2011). This definition of executive scandal requires that the misbehavior identified must involve the candidate (not incumbent candidates), a family member of the candidate, a senior campaign official, or the nominee’s running mate. The definition requires that the scandal was first revealed in the course of the campaign, whether the event happened in the past or recently.
To collect the information pertaining to each individual scandal (including the individual involved, the start of the scandal, the type of scandal, the allegations against the official), several sources were examined. These sources were typically secondary texts, following the lead of other scholars who have searched for a universe of political scandals (Kim and Bahry Reference Kim and Bahry2008; Maltzman et al. 2012; Puglisi and Snyder Reference Puglisi and Snyder2011). First, several newspapers were searched in multiple ways. By using newspapers and official sources to determine when scandals broke (and hence whether or not they existed), this minimized the presence of false or malicious rumors from the data. A series of search terms were used to capture individual scandals.Footnote 1 Using Lexis-Nexis Academic and searching by individual years, the search string was used to search Associated Press articles. Second, several excellent encyclopedic texts, which are compendia of scandals, corruptions, or ethics violations in government, were searched.Footnote 2 The span of these data include only the nomination period (i.e., until one of the candidates secures enough delegates to be designated the winner) and scandals involving nonincumbents (i.e., Whitewater for President Clinton and the Valerie Plame scandal for President Bush are excluded).
In establishing a time span for the models, the scandal begins when the charges are made public for the first time, and a scandal is considered ended when the accused individual is exonerated or formally leaves his or her position. The origin of the scandal can involve a news story in a major daily newspaper, a report to Congress, an internal investigation, or other secondary sources (see endnote 2 on sources). This variable spans the life cycle of the scandal from the first break of the story to the end of the scandal, whether that end be politically or legally favorable for the individual charged in the scandal. Each scandal has a definite temporal beginning but some scandals emerge and do not end. For instance, allegations might not be resolved by the time a campaign is over or another candidate wins the nomination. An “active” scandal (or scandals) is (are) indicated in the data by an integer registering the total number of scandals dealt with by an individual campaign on a given day (for instance, “1” for one scandal, “2” for two scandals).
Using the definition defended above, 54 scandals that occurred between the 1996 and 2012 election cycles were identified (beginning in January of 1996 and ending in November of 2012 with the end of the election cycle). Table 1 has a list of each scandal and the individuals associated with each scandal. These scandals span several personal, financial, and political issues. For example, in 1996, a top fundraiser for Senator Bob Dole entered a plea agreement to pay $6 million in personal and corporate files for illegally funneling campaign funds to the Dole campaign (Babcock Reference Babcock1996). In the 2008 election cycle, Antoin “Tony” Rezko, a longtime fundraiser for Senator (later President) Obama, was charged with fraud, attempted extortion, and money laundering (Drew and McIntire Reference Drew and McIntire2007). In 2012, Herman Cain acknowledged both new and old allegations of sexual harassment while he was the chief of the National Restaurant Association in the 1990s (Rutenberg and Shear Reference Rutengert and Shear2011). As a way to check the accuracy of these individual scandals, the list of events plus all resources—newspaper articles, books, and articles—were given to a research assistant, along with a rubric for deciding whether each event was a scandal. This second coder was assigned a subsample of scandals and asked to apply the same rules as the first coder. This process demonstrated significant agreement between the coders.Footnote 3
Table 1 Campaign Scandals, 1996–2012
NOMINATION RESOURCES DATA
In addition to the scandal data, several additional measures are used to capture key campaign resources or predictors of successful campaigns.
Fundraising Data
We use two sets of data from the Database on Ideology, Money in Politics and Elections (DIME) (see Bonica Reference Bonica2013). First, we chart the daily fundraising totals for each candidate or committee (Daily Fundraising Total). These data are compiled from contribution records, candidate and committee filings, and election outcomes from the Federal Election Commission.Footnote 4 Second, we summarize the ideological score of the contributors to each candidate or committee for each day (Donor Ideology). The DIME database has a “CFscore” for “direct distance comparisons of the ideal points of a wide range of political actors from state and federal politics” for more than 13.7 million individuals.
Candidate Vote Share (“Last Traded Price”)
To capture the relative candidate position in the race for the president, we use the lagged daily changes in the Iowa political futures market.Footnote 5 Higher traded prices reflect a market consensus that that candidate is more likely to win the nomination. These data are counted daily and are as accurate as opinion polling in predicting election outcomes. The share price ranges from $0.00 to $1.00. These market data also are consistently shown to be highly correlated with national election polling and with eventual political success (Erikson and Wlezien Reference Erikson and Wlezien2008). Thus, the independent variable for each model is the “last (closing) price” of that candidate as traded on the Iowa futures market, lagged by one day. We argue that this is a reasonable approximation of the relative probability that an individual candidate will win, perhaps more closely measuring daily political support than public opinion polls.
Endorsement Data
We use multiple measures to capture the number of endorsements on a specific day of each presidential candidate (Endorsements). First, we use endorsement data by party elites (Dowdle, Adkins, and Steger Reference Dowdle, Adkins and Steger2009; Steger Reference Steger2007). These data were measured from a content analysis of newspapers across the nation meaning they were public announcements of support, including governors, US senators, statewide party or elected officials, US representatives, state senators, state representatives, and local officials. These data are supplemented with data from Cohen, Karol, Noel, and Zaller (Reference Cohen, Karol, Noel and Zaller2008) who track endorsements with a “search of major national and regional publications, such as the New York Times, the Los Angeles Times, and the Chicago Tribune, plus any magazines that follow presidential politics.” The number of endorsements was summed by the candidate by the day for the 1996 to 2004 period to match the scope of the data.
Primary / Caucus Day
As a control variable to determine the effect of winning a presidential primary or caucus in a state on a particular day, a dummy variable is included to account for the presence of a candidate winning a primary or caucus (Primary Victory). This is coded “1” if the candidate won a primary or caucus on the day in question and “0” if there was not a primary or if the candidate did not win a primary or caucus that day. These data were taken from The Race for the Presidency (Cook Reference Cook2008), The Road to the White House (Wayne Reference Wayne2008), media reports, and The Green Papers website (www.greenpapers.com). A candidate winning an individual state event may cause a spike in perceptions of the ability of that campaign to secure the nomination and in fundraising (Corrado Reference Corrado and Johnson2014, 69), making them an important control variable.
DO SCANDALS HURT CANDIDATES SEEKING THEIR PARTY’S NOMINATION?
As discussed previously, when scandals emerge during the quest for the nomination of a party’s presidential bid, these events should have a negative impact on the candidates involved. But, how much of an effect is there? The results from a series of fixed effects panel OLS models (using the individual campaigns and election year as the unit) for the three key dependent variables, displayed in table 2, demonstrate significant but not uniform effects on key indicators of candidate success.Footnote 6 To begin, for the total daily funds raised, candidates affected by political scandal are significantly less likely to raise funds over the course of the life of the scandal. In particular, candidates involved in scandal raise approximately $40,000 less per day than candidates not involved in scandal. The last traded price also has a significant positive effect on fundraising. Candidates who are more likely to win tend to have greater ability to raise campaign funds. Republican candidates’ fundraising is more harmed by scandal given that the coefficient is larger for these candidates than for all candidates. On average, Republican candidates’ fundraising daily yield is about $20,000 less than when examining all candidates combined, about $60,000 less per day.
Scandals have a negative effect on funds raised but a positive effect on the final price of daily traded shares of individual candidates.
Table 2 Effects of Scandal on Fundraising, Position and Endorsements, 1996–2012
Note: A fixed effects panel OLS model was used for the analysis (set with election year and individual campaign set as the unit). The dependent variable are the total amount of funds received on an individual day, the last traded price on the Iowa Political Market on an individual day and the number of endorsements a campaign received in an individual day (‡ notes 1996 to 2004 because of data availability). Standard errors in parentheses. *** indicates statistical significance at p<.01 ** indicates statistical significance at p<.05. * indicates statistical significance at p<.10.
Scandals have a negative effect on funds raised but a positive effect on the final price of daily traded shares of individual candidates. The results in table 2 demonstrate that the more scandals faced by a candidate has a small positive effect on a candidate’s position for both all candidates and Republican candidates. That the effect is positive seems to challenge the notion of scandals negatively affecting a candidate’s ability to get the party nomination. Why? It may be that any publicity is good publicity. This may be especially true when considering what amounts to an open seat for the nomination (because the analysis only includes non-incumbents). In a crowded nomination field, as many of the cases examined here were, attention to an individual’s campaign may provide a temporary (and modest) boost to a candidate’s campaign. Some evidence suggests that a candidate’s duration in a race is a function more of their ability to raise their profile rather than raising funds or media coverage (Damore, Hansford and Barghothi Reference Danmore, Hansford and Barghothi2010). Other scholars have demonstrated that the effects of scandal on candidate approval is short lived (Miller Vonnahme 2014). Even so, the substantive results are small, ranging from .02 to .03 on a scale of 0 to 1 per day, making the added positive effect somewhat marginal.
Although there are significant effects for total funds raised and the last traded price of each candidate, there is no effect of scandal on the number of endorsements for all candidates. The effect of scandal on endorsements, as might be predicted, is negative but not statistically significant. Cohen, Karol, Noel, and Zaller (Reference Cohen, Karol, Noel and Zaller2008) report that some potential endorsees hold back endorsements of candidates in the aftermath of scandal (181). This effect does not appear significant when examining all candidates in primaries during this time period. The truncated nature of the data may be causing the limited effect. However, Republican candidates are more likely to lose endorsements than all candidates grouped together, generally as expected. The coefficient for Republican candidates in table 2 shows a negative statistically significant effect of scandals on endorsements, where Republican candidates lose between one and two endorsements per day with the presence of or an increase in the number of scandals associated with the campaign.
CONCLUSION
The range and depth of political scandals has both short- and long-term implications to governance. This article has extended these effects to the ability of aspiring presidential nominees to function using resources associated with successful capturing of a party’s nomination for the presidency. The changing nature of primary election politics necessitates an understanding of the factors that enhance or detract from a candidates’ ability to win in competitive primaries. Hagen and Mayer (Reference Hagen, Mayer and Mayer2000) argue that “the combination of early withdrawals and increased front-loading greatly accelerates the voters’ decision process and thus makes the whole system less deliberative, less rational, less flexible, and more chaotic” (40). To that point, we could add nomination scandals to this list. The history of Gary Hart’s ironically named “Monkey Business” boat (previewing other illicit behavior) to Bill Clinton’s “I did not inhale” defense against charges of illegal drug use are more than fodder for late-night talk shows. These political events have real, serious, and consequential ramifications for which candidate gets the nomination and the effectiveness of their campaigns.
The front-loading of the presidential primary season creates an environment where candidates must compete early and vigorously for the nomination of their party (Ceaser, Busch, and Pitney Reference Ceaser, Busch and Pitney2013). Viability and perceptions of efficacy are critical to candidates in these contests. From the evidence presented here, the most significant impact of scandal is on a candidate’s ability to raise campaign funds. Viability is signaled by the ability to increase one’s campaign coffers, especially in crowded fields where no candidate demonstrates the ability to take the lead in a race (Corrado Reference Corrado and Johnson2014, 67). Perceptions of vulnerability because of scandal clearly take a toll on individual candidates’ fundraising efforts. Oddly, scandals have a small positive effect on a candidate traded shares. Campaigns may aggressively defend themselves during these moments, causing a small increase in popularity (Greener and Arterton Reference Greener and Arterton2009, 179). Another explanation is that, in a crowded field, any publicity, even bad publicity, may help individual candidates receive higher name recognition than their opponents. For all candidates grouped together, there is no effect of scandal on the number of endorsements, suggesting that the decision to endorse among party elites is driven by factors not necessarily involving scandal, where scandals may not be strongly tied to perceptions of the skill to govern or ideological ties.
Republican nominees are more significantly disadvantaged by scandals during the race for the nomination of the Republican Party. The depth of this negative effect is most prominent in fewer funds raised and fewer endorsements garnered. Republicans may be more heavily chastised for hypocrisy in government (Basinger and Rottinghaus Reference Basinger and Rottinghaus2012). Rank and file members of the Republican Party, especially more conservative primary voters, are more critical of their potential nominee’s illegal or immoral actions because the voters and activists in the Republican Party are more conservative and adhere to more traditional values than other party members (see also Welsh and Hibbing Reference Welsh and Hibbing1997). Because the Republican Party has become more solidly conservative (McKee and Hayes Reference McKee and Hayes2009) and has focused more intently on social issues over business issues (Miller and Schofield Reference Miller and Schofield2008), primary voters may be less willing to vote and elites less willing to back nominees who have past or current ethical issues.
