Abstract
Political science research is conflicted about the impact of political scandals on survival in office. Scholars have found strong negative impacts to some scandals but others have found minimal or no effects. The literature has explored several consequences but no one work examines them collectively. This article examines presidential, gubernatorial, and Congressional scandals from 1972 to 2021 to assess the impact of scandal in a polarizing America. We find the negative consequences from scandals vary across time and institutions. Scandals in the Watergate era led to more resignations in Congress but fewer resignations of White House officials in the 1990s. During the Trump administration, White House officials did not survive in office at rates greater than past eras, demonstrating little support for the “Trump Effect.” However, politicians generally survived scandal more in the polarized era, hinting at the changing role of political scandals.
“I could stand in the middle of Fifth Avenue and shoot somebody, and I wouldn’t lose any voters, OK? It’s, like, incredible.” -- Donald Trump.
1
In a polarized world where affective polarization is high, trust in media is low, and partisanship shapes news attention, do scandals matter in politics anymore? One body of work suggests scandals do have negative political consequences. The type of scandal, the frame of the scandal in the media (Shah et al. 2002; Darr et al. 2019; Wolsky 2022), and even the attractiveness of the candidate (Stockemer and Praino 2019) all contribute to the political impact of a scandal. Allegations of wrongdoing leveled at elected officials at several levels can hurt them politically. Congressional investigations depress presidential approval ratings (Kriner and Schickler 2014) and investigations into Congressional members’ campaign finance violations hurt them in subsequent elections (Wood and Grose 2021). Fousek and Wasserman (2010) cite “the rapid decline of public support for presidents whose leadership appears ethically compromised.” A scandal might diminish a president’s ability to enact legislation (Andolina and Wilcox 2000; Newman and Forchimes 2010). A poorly handled scandal may also lead to policy paralysis (Meinke and Anderson 2001). In the long term, political scandals undermine the public’s trust in government, especially when a scandal involves political accountability (Miller 1999; Chanley, Rudolph, and Rahn 2000; Barnes, Beaulieu, and Saxton 2018).
Another body of scholarship, however, suggests that in a polarized political world, scandals may not have a large effect on incumbent political officials. A polarized America means scandals matter less to partisan voters. Scandals aren’t necessarily “dealbreakers” for many voters (Funck and McCabe 2021). Williams (1998) laments that “there is no obvious correspondence between the degree of controversy generated by scandals and the gravity of the alleged misdeeds. Some of those involved in scandals pay a heavy price: resignation, disgrace, and even imprisonment. Others, who seem equally culpable, escape conviction and retire with dignity” (2). Specifically, partisanship limits the negative impact of scandals (Entman 2012; Dziuda and Howell 2020; Wolsky 2022; Funck and McCabe 2021). Partisans stress different messages about scandals based upon partisanship (McKee, Evans, and Clark 2021) and may rake in additional campaign contributions from party faithful (Hamel and Miller 2018). More media coverage may also simply generate “white noise” that the public ignores (Lawrence, Bennett, and Hunt 1999) or ambivalent reactions that “all politicians do it” (Tiffen 1999).
Scandals affect politicians, political parties, and the institutions they serve (Bowler and Karp 2004) so understanding the impact of scandals on elected officials helps us understand the foundations of democratic rule. Markovits and Silverstein (1988) note “political scandals represent a challenge to the very legitimacy of liberal democracies,” making it critical to study the institutional effect of scandals (10; Quirk 1998; Sabato 2000; Sabato 2007; Basinger and Rottinghaus 2012). Using three new data sets of scandals involving presidents and members of Congress at the national level and governors at the state level from 1972 to 2021, we chart the duration of each political, personal, and financial scandal faced by an elected official. We then investigate what factors quicken the “end” of the scandal, defined as when the scandal ends in a negative outcome for the individual. These results help clarify the conventional wisdom of how politicians survive (or do not) scandals while in office and whether these trends are exacerbated by the political climate of several distinct eras of political scandals.
When Scandals Do (and Don’t) Matter
Scandals have given an avenue to the venom in partisan politics for a generation since Watergate (Ginsberg and Shefter 1999). Scandal is one facet of modern contentious American political life and one factor in the growth of the “public integrity war” (Roberts and Doss 1997). Scholarship demonstrates that political scandals can negatively impact political careers. Lazarus (2008) showed that scandal-tainted incumbents were more likely to face opposition in both their primary and general elections and were more likely to face challengers with electoral experience. This effect is pronounced for minority candidates who face additional scrutiny (Rajan and Pao 2022). Politicians are reelected at lower rates following scandals (Welch and Hibbing 1997; Praino, Stockemer, and Moscardelli 2013; Basinger 2013). This is in part because scandals often strangle political fundraising (Rottinghaus 2014). Scandals have a minimal effect on legislation at the national level but leads to more friction between the branches and more presidential vetoes as the number of scandals in an administration mount (Rottinghaus 2015). Political cynicism from current scandals towards elected officials affects how individuals interpret information about future political scandals (Dancey 2014).
But, while some scandals are a liability, others are not (Miller 1999; Zaller 1999; Busby 2022). Nyhan (2009) found that there were more scandals when government was divided, suggesting a polarized effect of scandals that are weaponized politically. Scholars have also demonstrated that the effects of scandal on candidate approval are short lived (Vonnahme 2014) and fade quickly (Pereira and Waterbury 2019). Voters rate scandal-plagued politicians more negatively than other politicians but still prefer ideological proximate candidates in terms of partisanship and shared political views even considering the emergence of a scandal (Funck and McCabe 2021). Perception of politicians and their susceptibility to corruption also affects the impact of scandals. Barnes, Beaulieu, and Saxton (2018) find people are less skeptical of politicians’ misdeeds when there is a greater potential for institutional sanction (see also Barnes and Beaulieu 2014; Saxton and Barnes 2020). The type of scandal matters as well. People respond more negatively to financial scandals than to moral scandals when they do not involve abuses of power (Doherty, Dowling, and Miller 2011; Doherty, Dowling, and Miller 2014). Corruption and sex scandals tend to be more damaging to reelection prospects than political scandals (Basinger 2013).
If scandals are a breach of the public trust, why do some not have a strong negative political impact? Changing political circumstances insulate some politicians from the deleterious effects of political scandals. In a polarized world where affective partisanship is high, trust in media is low, and people live in their own media ecosystems, the impact of scandal may not have the same negative political effect it used to have. Presidents and their cabinets and governors survive scandals more frequently when they have greater partisan support, reducing the potential of removal from office (Rottinghaus 2014). The rise of beliefs in conspiracy theories muddies what the truth is for some Americans (Uscinski 2020). Voters are now more likely to vote for candidates who promise to pass policies that “disproportionately harm” supporters of the opposing political party (Webster, Glynn, and Motta 2022). Partisan media polarize people (Levendusky 2009) and media distrust leads voters to discount certain kinds of information, especially when it is negative news about a co-partisan politician (Darr et al. 2019). Media coverage of scandals is partisan and shaped by ideological leanings of the media (Galvis, Snyder, and Song 2016; Puglisi and Snyder 2011). Beyond partisanship, the Supreme Court’s loosening of bribery and corruption laws make it harder to prosecute and convict an elected official (Teachout 2017), giving some politicians a legal shield to accompany their rhetorical sword.
A Polarized America Mutes the Impact of Scandals
Do scandals still matter in a polarized political environment? The literature is mixed on the impact of political scandals but leads to one clear expectation regarding the impact of modern scandals on elected officials: politicians in contemporary political times are more likely to survive scandal. Explaining then testing this expectation occupies the rest of the article.
Why? Partisanship has hardened political preferences. Scandals affect partisan identifiers differently than others. For both scholarship on hypothetical and real life cases, partisans believe scandals involving in-party politicians are less serious or problematic than those committed by out-party politicians (Anduiza, Gallego, and Munoz 2013; Walter and Redlawsk 2019; Costa, et al. 2020). Even in the face of clear and undisputed charges, co-partisans backed a hypothetical scandal-plagued president despite evidence of illegal and “impeachable” activities (Cortina and Rottinghaus 2017). Partisans weigh scandalous information about out-party politicians more heavily than the same information about co-partisans (Goren 2002). Even in facing serious charges of sexual harassment, Republican partisans do not strongly penalize their own political candidates (although Democrats are more likely to penalize their own candidates) (Stark and Collignon 2021). While scandal-tainted politicians generally receive fewer votes and raise less money, when partisan politics emerges in elections, voters are less punitive of scandal-ridden politicians (Rottinghaus 2014). Politicians may simply be better at managing a scandal as a crisis than in earlier periods (Busby 2022).
Excessive partisanship and perceived threats from opposition parties amplifies this rigid, asymmetric effect (Woessner 2005; Simonsovits, McCoy, and Littvay 2022). Rising polarization between the two parties increases the likelihood of a scandal, since both sides are looking for a political edge, while also decreasing the importance of that scandal to a partisan audience (Dziuda and Howell 2020). Voters are less punitive and donors are often even more supportive of candidates facing scandals in the post-1994 period of nationalized elections (Hamel and Miller 2018). These factors are exacerbated since then. Ginsburg and Shefter (1999) conclude that partisanship increases scandals as the “perfect nonelectoral weapons” even while they fail to mobilize voters to vote violating politicians out of office. Even presidential approval, normally sensitive to changes in the economy, now has an attenuated relationship where presidential approval “appears to be a proxy for partisanship” (Small and Eisinger 2020; see also Lebo and Cassino 2007). The media neglect to cover most scandals like corruption (Entman 2012) and, when they do, the coverage is often slanted by the ideological leanings of the individual media organization (Puglisi and Snyder 2011).
A predominant factor in predicting survival of scandal is partisan support, where, for instance, greater numbers within the politician’s political party provide for “pillars of support” (Shear 2011; Miller 2010). Specifically, surviving scandal is intimately linked to the ability of an elected official to stay in office. One element of this is related to the likelihood of removal from office for the crimes associated with the scandal (Baumgartner and Kada 2003). For instance, the capacity to stay in office is directly related to the amount of support a chief executive might have, where more partisans in a legislature is likely to mean a greater chance of political survival (Dimock and Jacobson 1995). Put another way, more partisans create a legislative shield for a chief executive (Hinojosa and Perez-Linan 2006). A chief executive’s fellow partisans are less likely to vote to impeach or remove a chief executive from office and can therefore be counted on as a measure of political strength during a scandal. For Members of Congress, unified government decreases the likelihood of internal investigations through the House Ethics Committee which can lead to negative electoral consequences (Basinger 2016).
Methods and Data
Defining Scandal
Several definitions of political scandals exist. For instance, Marion (2010) requires that a public figure has been “accused of unethical or immoral behavior” defined as offending behavior or an event “that is disgraceful, shameful, or discredits someone” or that transgresses “societal norms, moral codes, or values” (11; Genovese and Farrar-Myers 2012). Thompson (2000) offers a detailed definition which requires that actions “transgress or contravene certain values, norms or moral codes” and that the actions’ disclosure might damage responsible individuals’ reputations, so that they attempt to conceal the action. Our definition requires that an individual scandal must involve allegations of illegal, unethical, or immoral wrongdoing. The definition includes adultery because of the unique place of inappropriate sexual relationships in the political world (Rosen 2009). Including sex scandals ensures that we include “transgressions” of conventional morality, but we wish to exclude gossip, innuendo, and unsubstantiated rumors of private behavior. 2
Our definition of executive scandal requires that the misbehavior identified must involve the president, a member of Congress, a governor, the immediate families of these politicians, a senior administration official, or a federal nominee. The unit of observation is the individual involved in the scandal. Specifically, at the federal level, we include scandals involving the vice president, cabinet secretaries, officials with cabinet level rank (the White House Chief of Staff, or the Director of the Office of Management and Budget), agency heads at the federal level, the elected officials’ family, and senior campaign staff. 3 We also include nominees for national and state executive- or judicial-branch positions during the period when a nomination is under consideration. At the state level, we include scandals involving the governor, senior staff (including agency heads and high-level political appointees), the immediate family of the governor, and senior campaign staff. The scandal had to take place during the individual’s time in office (not, for instance, revelations after the principal or staff member left office). For comparison, we have also included an updated dataset on scandals involving members of Congress (Basinger 2013; Hamel and Miller 2018).
Using the definition defended above, we identified 156 presidential level scandals (involving 198 individuals), 338 gubernatorial level scandals (involving 385 individuals), and 327 scandals involving members of Congress that occurred between 1972 and 2021 (ending in January of 2021 with the end of the Donald Trump term). 4 Using secondary texts follows the lead of other scholars who have searched for a universe of political scandals (Kim and Bahry 2008; Puglisi and Snyder 2011). If the principle focus of the scandal was the elected official, we recorded this as well and included it in the model to control for the fact that the duration of the scandal will be different for the elected officials than others, for example, the president versus an agency head, as discussed above. This also allows comparable evaluation of political figures, at both the state and national level, who have unique institutional settings. An online appendix uploaded to an article page in the Harvard Dataverse has a full list of the scandals and the individuals associated with each scandal.
Type of Scandal
For purposes of analysis, we focus on two categories: financial scandals (Financial Scandal) and personal scandal (Personal Scandal). Financial scandals are scandals where an individual personally financially profits from his or her actions, such as embezzlement of funds, accepting bribes or political payoffs, and non-payment of state or federal taxes (Thompson 2000). For instance, in 2019, Republican Representative Chris Collins was arrested by the FBI, charged, and plead guilty to wire and securities fraud for tipping off his son with insider financial information. Personal scandals involve the immoral or unethical personal behavior of an individual, especially adulterous in nature (Lewinsky affair during the Clinton Administration) but also including use of illegal drugs (Supreme Court nominee Douglas Ginsberg), illegal household staff (Secretary of Labor nominee Linda Chavez), or theft (see Kaygay 1999). The third group (not included) is political scandal.
Institutional Variables
To examine the effect of institutional variables, we examine the relationship between the executive and legislative branches—which may lead to removal of office for an official—through two variables. First, we include a dichotomous measure of divided government, where any amount of divided government (in either or both chambers at the federal or state level) is treated as divided government (Divided). Second, we include variables for the number of co-partisans in both the upper (President’s or Governors’ Partisans Senate) chamber in the same party as the president or governor. 5 National level data were taken from Vital Statistics on American Politics (2011). State data are updated from Klarner (2003). 6
Polarization and Scandal Era
Since polarization is a contentious (Fiorina and Abrams 2008) but continuous process, we use a dummy variable to indicate the pre-and-post polarized era cut in 1993 with the election and inauguration of Bill Clinton, a period of rising elite political polarization (Layman, Carsey, and Horowitz 2006) and “weaponization” of scandal (Sabato 2000). Although scandals have beset politicians since the beginning, scandals moved to “institutional combat” in the 1990s as polarization grew (Ginsburg and Shefter 1999). In addition, to explore whether the negative effects of scandals are worsening or improving over this time period, we specify the span of the full data into decades, which roughly correspond to the major scandals of that era: the 1970’s (Watergate and post-Watergate; n = 48), 1980’s (Iran-Contra, House banking scandal, and Abscam; n = 146), the 1990s (the Monica Lewinsky scandal; n = 215), the 2000s (Abramoff scandal; n = 250), and the 2010s (multiple governors and Donald Trump scandals; n = 249). These time divisions encompass multiple unrelated scandals but allow for a rough estimate of the impacts of scandals in each era. As expected, there are differences across time in survival rates of politicians in each era: 1970’s (46% resigned or removed from office), 1980’s (52%), 1990s (40%), 2000s (45%), and 2010s (44%).
Control Variables
To examine any temporal aspects of the survival of scandal, we include two key control variables in the models below. First, the effects of scandals on politicians in office is gendered where women politicians involved in scandals activate traditional gender norms and provoke more hostile respondents to punish women more severely than men (Barnes, Beaulieu, and Saxton 2018; Armstrong et al. 2022; Courtemanche and Green 2020). To capture this dynamic, we include a control variable (coded “1”) for the gender of the individual (Female). Second, to measure executive approval at both the national and state level, we recorded the approval measure of the president or governor taken as close to the date of the break of the scandal story (but not after the story broke) as possible (Approval). The question asked “Do you approve or disapprove of the way [President or Governor (name)] is handling his job as [(president) or (governor)]?” Responses are collapsed into positive and negative values, excluding those who responded “don’t know.” National level data were taken from iPoll’s report of Gallup Poll data for the appropriate time period. The state-level data was taken from the U.S. Officials’ Job Approval Ratings (JARs), a Cooperative Project of the University of Rochester, the University of North Carolina at Chapel Hill and George Washington University. 7 In instances where more accurate state-level data could be located at Survey USA, those data were used. 8 We also included a variable (House) to distinguish between House members and Senators only in the Congressional models. Finally, we capture party by coding whether a politician was a Republican or not (Republican).
Model
In that we are interested in the factors that contribute to a politician’s “survival” of a scandal, we use a series of duration models. 9 A Cox proportional hazard model is a semiparametric model that allows the hazard model to be unestimated—this is especially useful when the shape of the baseline hazard model is unknown (Cleves, et al. 2010). This model also assumes the shape of the hazard is the same for all the subjects, a reasonable assumption considering that the effect of scandal in the modern era (especially post-Watergate) should be theoretically similar. The Cox model asserts that a hazard rate for the jth subject in the data is h(t| xj ) = h 0 (t) exp(x j β x ), where the regression coefficients, β x , are estimated from the data (Cox 1972). This also allows us control for duration dependence (Box-Steffensmeier and Jones 2012; Zorn 2000). The parameterized effects of each variable are measured with the variables below using categorical and continuous variables. We also cluster by scandal since the effects of each scandal (including the severity and personnel involved) are likely to be similar within each scandal, providing within-subject correlation (Lin and Wei 1989). This provides a valid representation of the “sample-to-sample variability” of the coefficients (Cleves, et al. 2010). Although selection effects are problematic for some duration models (Boehmke, Morley, and Shannon 2006), the nature of the data collection for these data resolves these possible issues. 10
Scandal Length and “Failure”
In establishing a span of time for the models as required for duration models, the scandal begins (t1x) when the charges are made public for the first time and the and a scandal is considered ended (tnx) when the accused individual is exonerated, formally leaves his or her position, or the presidential or gubernatorial administration ends. The origin of the scandal can involve a news story in a major daily newspaper (searched in Lexis-Nexis), a report to Congress, an internal investigation, or other secondary sources (see above for note 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 and end, allowing for completion of the duration models.
To be more specific as required for predicting the end to the span of the sequence, “failure” in the duration models (or the right censored variable) below is identified as when the scandal ends in a negative outcome for the individual or the administration. This includes the following scenarios: an individual being fired from their position, resigning (or being forced to resign), being indicted at any level or removed from their position or office in some way. Failure in this case excludes instances where an individual is charged with a crime after they leave office or they are indicted but remain in office until their term runs out. Non-failure, such as evidence of an extra-marital affair that lasts the duration of the incumbent’s time in office, does not have an official end so the time length spans the rest of the incumbent’s time in office. 11 If a politician is exonerated, including evidence of innocence, a favorable court ruling, a committee report with evidence of innocence, or the guilt of another associate, or restitution (a fine paid), the scandal is considered ended.
What Increases the Risk of a Scandal Ending Badly for Politicians?
Do different eras of scandal independently impact survival of scandal? Figure 1 graphs the rate at which scandals end badly for each office across five eras (lower percentages mean more survival). The results show that politicians generally survived most frequently in the 1990s. For most of the major scandals during the time period, White House officials (including President Clinton himself), governors (George Ryan of Illinois and Edwin Edwards of Louisiana), and members of Congress faced political scrutiny but revelations of scandal did not always lead to resignations or removal from office (or the effect was delayed considerably). Survival percentages for the other eras were mixed across office type. Survival of state executive scandals was consistent across time, although slightly higher in the 1970s. For federal executives in the 1980s and 2000s, more officials left office due to scandal than other periods. The 2010s had higher than average rates of survival than the average for across all institutions. Some have labeled this the “Trump Effect” and point to the former president for a changing nature of political scandals (“the ringleader of looser standards”) (Miller 2019). Earlier periods of scandal history (the 1970s, the Watergate era) had bad ending rates significantly higher than the other eras, especially for members of the House, implying politicians of earlier eras faced at least some more scrutiny and pressure to leave office than politicians in more contemporary eras. This may be because of saturated coverage of scandal during the era and fewer media sources to dilute coverage (Busby 2022). Scandal “end in disgrace” rates by era and office type.
Scandals Duration Models.
Note. Coefficients are exponentiated coefficients of the hazard ratio, clustered by scandal. Dependent variable: total time (in days) from when the scandal broke to when the scandal ended. Failure means the resolution to the scandal ended in resignation, firing, or removal from office. The standard errors are robust standard errors, clustered by each scandal. *** indicates statistical significance at p < 0.01. **p < 0.05. *p < 0.10. Robust standard errors in parentheses.
Table 1 identifies four models of scandals involving members of Congress, governors, presidents, and two reduced form (“combined”) merged models with all three office types together. 14 In the models for governor and president, if the scandal affected the elected official themselves (the elected governors or presidents), the individual was much more likely to survive the scandal. 15 This is not surprising since administrations confronted with scandals frequently rally around their elected leader to protect their greatest political assets (Rottinghaus and Bereznikova 2006). Party organizations and partisans often politically back their elected officials during scandal to insulate them from removal from office. House members (signaled by the “House” dummy variable) are significantly more likely to have a scandal end badly than a Senator (400% more likely). Female incumbents are less likely to face risk of a scandal ending a political career than males (more than 185%); however, this is only true for members of Congress, not federal or state executive officials. This finding confirms what others have found about the double standard for women involved in political scandals (Barnes, Seaulieu, and Saxton 2018), primarily in Congress. The relatively few number of women involved in scandal in executive offices may partially explain these modest findings.
The type of scandal also affects survival. Personal scandals hasten a quicker end to a political career for the combined models and significantly for members of Congress (432% more likely). Financial scandals end careers for presidential officials but not necessarily for other institutions. Partisanship matters too. Although there are no significant differences between Republicans and Democrats across the models, we do find a small statistical effect suggesting Republican governors survive more than Democratic governors (about 28% more likely). Surprisingly, the number of co-partisans in upper chambers (the institutional responsible for putative impeachment) and the presence of divided government has inconsistent effects. More co-partisans in a position to save officials only have strong statistical effects on scandal survival in the presidential models (about 8%). Divided government has a positive effect on survival for the two merged models for related reasons: without clear consensus on a political course of action as often occurs during divided government, attempts to impeach may fail.
Because of the growing partisan nature of modern politics, later era scandals should less frequently end political careers, as explained above. For the combined models, the 1990s and 2010s produced significant survival rates, confirming expectations that they would be more likely to survive in this polarized era. Figure 2 displays predicted survival (Kaplan-Meier) estimates for the pre-polarized (1972–1993) and post-polarized era (1993–2021) (results from the combined model in Table 1). Following the modest findings for individual offices in Table 1, the results in the graph show that survival rates in the polarized era are slightly higher than those of the prior era (about 23%), providing support for the expectation of polarized politics leading to scandal survival.
16
For individual offices, the expectation is not uniformly supported across political positions, although there is a significant survival rate for federal executives, cabinet, and staff during the time around the Lewinsky scandal. In contrast, scandals in the Watergate era for members of Congress leads to significant and large hazard rates, more than double those rates of later eras (more than 500%). Predicted smoothed hazard estimates by era.
Indeed, in Table 1, there is no consistent statistically significant survival effect for time periods across individual office types. In particular, presidential scandals were more likely to be survived in the 1990s than in the 2000s and 2010s (technically compared to the 1980s). The Clinton Administration weathered several scandals, most prominently the Monica Lewinsky affair which led to his impeachment but also including fallout from the Whitewater land deal, illegal foreign fundraising, the White House travel office controversy, and other allegations of sexual assault by multiple women. However, for presidents in this latter period, President Trump, Trump White House officials, and their staffs did not survive in office at rates greater than other White Houses past (although Trump as president did survive as expected). This finding for the 2010’s era contradicts assertions that scandals matter less than in prior eras. In fact, scandals seem more damaging to politicians. For instance, although several members of the Trump administration caught in scandal did not resign, including the president himself, others like Environmental Protection Agency head Scott Pruitt and Health and Human Services Secretary Tom Price resigned rather quickly. However, collectively there is a strong survival effect during the Trump era although not as strong as during the Lewinsky era. That the normally heighted survival rates for past White Houses (such as the Lewinsky era) was diminished during the Trump administration does not definitively support the assertion that the negative effects of scandals disappeared during the Trump era.
Discussion
Not all politicians survive scandal—every year scandals fell politicians of every stripe. Voters can still spot scoundrels, especially when opponents point out ulterior motives for a politician’s actions (Genovese and Morgan 2012; Robison 2021). But some elected officials survive scandals even when conventional wisdom suggests these scandals might be career-ending vulnerabilities. Indeed, demonstrating modest support for expectations, later politicians generally survived in office longer than earlier politicians. Later politicians, especially beginning in the aftermath of the Lewinsky scandal and continuing through the Trump era, collectively confronted hardened partisan preferences and this modestly inculcated them from the negative effects of scandal. But for some politicians, stronger opposition armed with ammunition to politically hurt incumbents in more ways minimizes the chances of survival. For instance, the political system appeared to “clean house” during the George W. Bush and Donald Trump periods where White House officials (but not the president) resigned or were fired at higher rates following major scandals.
The findings here also demonstrate institutional differences in survival of scandal. Senators (as opposed to House members), governors, and presidents are all individually likely to survive scandals. More frequent electoral cycles and the emergence of strong candidates at the local level to run for the US House amplify the negative effects of scandals on members of Congress. Senators, governors, and presidents face longer electoral horizons where the negative effects of scandals may deteriorate with time. State and nationwide officials may also have substantial significant political support and a more sizeable financial cushion to absorb political blows from opponents following scandals. Not surprisingly, the elected officials themselves (especially when the principles are governors and presidents) are the most likely to survive these scandals because of the protection provided by their political organizations and allies. This is especially true for incumbents in polarized times who reason that the likelihood the truth will emerge (and be believed) is lower since partisan media attention shapes perception and politicians may be more likely to prevaricate or stonewall (Basinger and Rottinghaus 2012; Basinger and Rottinghaus 2012).
When controlling for the individual time eras, neither divided government nor the number of co-partisans in the upper legislative chamber significantly affects the prospects of a political career post-scandal across institutions. For members of Congress, divided government could lead to more investigations into wrongdoing related to scandal which often leads to more resignations (Basinger 2016). For governors and presidents (and their administrations), divided government may be a signal of opposition party desire to use scandals as a weapon against out-partisan incumbents. Yet polarized politics seems to have generally changed these expected dynamics. Decision making about resignations (or terminations for staff) are done somewhat independent of the structural, institutional elements of the system. Or, put differently, partisanship has over time hardened the resolve of politicians to stay in office because of a near impossibility of removal as a political outcome or a political boost from fighting charges with partisan return fire.
Conclusion
Partisan political change and polarized fragmentation of the media landscape has created divisions in American politics even on moral or ethical transgressions involved in scandals. The literature is divided about the effects of scandals—some arguing that scandals negatively impact politicians while others suggest polarization minimizes the deleterious effects of scandals on incumbents. This article uses five decades of comprehensive data on scandals addresses this debate and finds political scandals still matter but the effects vary across institutions. Politicians in earlier eras faced greater heightened negative consequences of scandals than those of later eras. For members of Congress specifically, the negative consequences of scandals remained high in the Watergate era but it dissipated with time. For presidents, typically robust survival of scandal while in office has diminished since the Lewinsky era, despite a polarized political world protecting partisan politicians. Governors and their staff are relatively immune from these political trends even as partisanship seeps into state-level politics.
These results suggest modest support for the notion that partisanship reduces the negative impact of scandal on some incumbent politicians. The findings also underscore the unique dynamics for each institution and highlight potential differences between elected officials (included in the analysis) and those running for office (not included in the analysis). From an institutional perspective, the literature explains differences in political outcomes for specific politicians and events. This article expands on these findings and shows the effects of scandals are not singular but are instead institutionally diverse. Discussion of the impact of political scandals should assess the unique institutional arrangements of each political figure as well as the politics of the time in which they serve.
Supplemental Material
Supplemental Material - Do Scandals Matter?
Supplemental Material for Do Scandals Matter? By Brandon Rottinghaus in Political Research Quarterly
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
