Introduction
Political parties are increasingly going negative in their campaign advertising and election messaging (Geer, Reference Geer and Geer1998; Kaid and Johnston, Reference Kaid and Johnston1991; Krupnikov, Reference Krupnikov2011). At the same time, it is clear that party leaders and individual candidates are becoming increasingly relevant to considerations of vote choice and to the electoral success of political parties (Aarts et al., Reference Aarts, Blais and Schmitt2011; Bittner, Reference Bittner2011; Gidengil et al., Reference Gidengil, Blais, Nadeau and Nevitte2000; Johnston, Reference Johnston and King2003; Poguntke and Webb, Reference Poguntke and Webb2005). Clarke and colleagues, for example, refer to party leaders as “the superstars of Canadian politics” (Reference Clarke, Jenson, LeDuc and Pammett1991: 89). What is less clear, however, is the relationship between these two simultaneous trends in electoral politics. In this research note we are particularly interested in the targets, or objects, of negative campaigning, especially from the perspective of personalization. That is, is it opposing political parties or their leaders who are targeted in routine election campaign communication?
The note begins with a discussion of negative campaigning and personalization. We bridge these two separate literatures by developing the concept of negative personalization. While the negative campaigning literature has witnessed tremendous growth in recent years, the target of campaign negativity has not been fully explored. Likewise, although scholars are increasingly writing about the personalization of politics, this literature has not considered whether parties can personalize their opponents by focusing their messaging and attacks more on individual leaders than the parties they lead. Building on this literature we suggest a new concept: negative personalization. Negative personalization, as we define it, is an emphasis on opposing party leaders in campaign communication more so than on the parties that they lead.
Drawing on data from recent elections in Canada's largest province, we provide a preliminary empirical look at the dynamics of negative personalization in election campaign material. We do so by examining 53 television advertisements as well as more than 350 party press releases in order to gauge the target of negative party messaging (that is, the party or leader). Additionally, we take a closer look at the campaign dynamics that shaped negative personalization during the 2011 and 2014 Ontario provincial elections.
Based on this preliminary analysis, we find that negative personalization is a relatively common feature of election campaigns and that the targets of negative personalization are typically unpopular party leaders. While the conclusions presented here need to be tested in other cases before they can be generalized beyond this particular analysis, this note does offer the first theoretical and empirical look at the concept of negative personalization as well as provide suggestions for future research.
Personalization and Negative Campaigning: Bridging the Gap
Negative campaigning is defined as “criticizing the record of the opposing party or parties; questioning the judgment, experience and probity of opposing leaders; and generating fear about what the future might hold if the opposing party or parties were in power” (Sanders and Norris, Reference Sanders and Norris2005: 526; see also Skaperdas and Grofman, Reference Skaperdas and Grofman1995). As Damore (Reference Damore2002) suggests, much of the work on negative campaigning has examined its prevalence (Buell and Sigelman, Reference Emmett and Sigelman2008; Fowler and Ridout, Reference Fowler and Ridout2013; Geer, Reference Geer2006; Kaid and Johnston, Reference Kaid and Johnston1991; Krupnikov, Reference Krupnikov2011; Lau and Pomper, Reference Lau and Pomper2004) as well as its effects on voter turnout, candidate evaluations, political knowledge and vote choice (see, for example, Brians and Wattenberg, Reference Brians and Wattenberg1996; Kaid, Reference Kaid1997; Wattenberg and Brians, Reference Wattenberg and Brians1999). Notwithstanding a recent meta-analysis revealing that negative campaigns do not achieve their desired effect (Lau et al., Reference Lau, Sigelman and Rovner2007), negative advertisements remain a common feature of modern election campaigns.
In fact, according to Lau and Rovner, “One of the most important decisions candidates make is whether to run on their own merits—that is, their own policy ideas, past accomplishments, and personal strengths…or if instead their campaign will concentrate on the perceived weaknesses of their opponent's policy proposals, prior policy failures, and/or personal peccadilloes” (Reference Lau and Rovner2009: 286; see also Damore, Reference Damore2002: 670). A body of literature has emerged focusing on this very question: why and when do candidates and parties go negative? The existing research suggests that competitive races tend to be more negative (Lau and Pomper, Reference Lau and Pomper2004); candidates who are trailing in electoral support are the most likely to go negative (Skaperdas and Grofman, Reference Skaperdas and Grofman1995); and leading candidates are the most likely to be attacked (Haynes and Rhine, Reference Audrey and Rhine1998).
However, strategic decisions do not end with the decision to go negative. By its very definition, negative campaigning can take on a variety of different forms and targets. While the definitions outlined above clearly include both parties and leaders and candidates, the literature has not typically separated the two. As Hansen and Pedersen (Reference Hansen and Pedersen2008) note, these definitions, as well as much of the empirical literature, have emerged out of the American context and tend to focus on candidates (who are treated synonymously with their party). A finer distinction between the various targets, however, is necessary, especially when considering multi-party competition outside of the United States. Who are the primary targets of campaign communication: parties, leaders, or both? Moreover, do parties in multi-party competition treat their various opponents differently in a strategic manner (that is, targeting some opposing leaders but not others)? These questions are particularly relevant in parliamentary democracies where parties have typically been the central players.
Students of political parties and elections have identified an increasing trend in recent decades towards more candidate centered politics (Cross and Young, Reference Cross and Young2015; Karlsen and Skogerbo, Reference Karlsen and Skogerbø2013; McAllister, Reference McAllister2015; Wattenberg, Reference Wattenberg1995; Zittel, Reference Zittel2015), individualized local campaigns (De Winter and Baudewyns, Reference De Winter and Baudewyns2015; Eder et al., Reference Eder, Jenny and Muller2015; Zittel and Gschwend, Reference Zittel and Gschwend2008), and a personalization of politics more generally (Balmas et al., Reference Balmas, Rahat, Sheafer and Shenhav2014; Rahat and Sheafer, Reference Rahat and Sheafer2007; Poguntke and Webb, Reference Poguntke and Webb2005). These trends point to changing electoral and political norms in which the centrality of individual actors has increased while emphasis on the political party has declined (see Rahat and Sheafer, Reference Rahat and Sheafer2007: 65; Karvonen, Reference Karvonen2010:4).
The rise of personalization is said to result from recent institutional changes as well as long-term societal transformations in voting behaviour. The adoption of primaries (Rahat and Sheafer, Reference Rahat and Sheafer2007; Ware, Reference Ware2002), the introduction of televised debates (Reinmann and Wilke, Reference Reinnemann and Wilke2007: 109) and the predominance of the horse-race frame (Taras, Reference Taras2001) have all contributed to the personalization of election campaigns. Empirical studies of the UK (Foley, Reference Foley2000), Israel (Rahat and Sheafer, Reference Rahat and Sheafer2007), Germany (Schulz and Zeh, Reference Schulz and Zeh2005), the United States (Dalton et al., Reference Dalton, McAllister, Wattenberg, Dalton and P2000), Canada (Cross et al., Reference Cross, Malloy, Small and Stephenson2015; Mendelsohn, Reference Mendelsohn1993, Reference Mendelsohn, Holmes and Taras1996), and a number of other countries confirm the increasing importance of candidates and leaders during election news coverage.
On the societal front, many Western democracies have experienced a significant decline of partisanship (Dalton and Wattenberg, Reference Dalton and Wattenberg2000) and increased electoral volatility (Bartolini and Mair, Reference Bartolini and Mair1990; Caramani, Reference Caramani2006). As the importance of historical and long-standing partisan cleavages and attachments declines, parties are increasingly turning to their leaders and candidates to win “personal” votes. At the same time, voters are evaluating these individuals and taking their characteristics into account when making decisions at the ballot box (Bittner, Reference Bittner2011).
Notwithstanding all of these changes, parties themselves still have a role: they can choose to resist the trend towards personalization or to embrace it. Adam and Maier capture the strategic element of personalization: “If a candidate is more popular than his/her own party, the attempt to transfer this positive image to the party by focusing on the candidate makes sense” and “if a candidate is significantly more popular than the opposite candidate, the party will, of course, try to build on this advantage” (Reference Adam, Maier and Salmon2010, 237). Personalization is therefore not only the result of changing voter behaviour, electoral institutions and media coverage but also of strategic intra-party decisions. These strategic elements, however, have rarely been considered by the personalization literature.
Despite its own strategic desires, a party that chooses to focus on its overarching brand instead of its leader may nonetheless experience personalization in the media and in the perceptions of the electorate. Largely ignored in the literature is the external factor, that is, other political parties. We suspect that personalization is not only a strategic intra-party decision but that there is also an inter-party element. Just as individual parties need to decide whether to focus on their own party brand or leader, they also decide whether to focus their attacks on competing parties or their leaders.
If party leaders are in fact becoming increasingly important to the decision-making calculus of voters and to election outcomes (Aarts et al., Reference Aarts, Blais and Schmitt2011), we expect parties to react to this by making strategic decisions regarding the attention they direct at rival party leaders. While there is a growing literature concerning parties’ strategic decisions regarding negative campaigning (Damore, Reference Damore2002; Krupnikov, Reference Krupnikov2011; Peterson and Djupe, Reference Peterson and Djupe2005; Walter and Vliegenthart, Reference Walter and Vliegenthart2010), none of this work centres on the degree to which parties choose to focus on opposing leaders rather than opposing parties.
It is therefore necessary to distinguish between positive and negative forms of personalization. Despite all of the scholarship examining how the media report on leaders, how voters respond to leaders, how candidates and leaders portray themselves in relation to their party and how parties negatively campaign against their opponents, we are unaware of any studies that examine personalization from this strategic and inter-party perspective.
Hypotheses
We investigate the possibility of negative personalization in the remainder of this research note through an examination of party-sponsored campaign messages and literature. In particular, we explore the extent of negative personalization as well as the conditions and circumstances under which party leaders are most likely to be the targets of negative personalization. While exploratory, we have developed two hypotheses and offer a number of possibilities for future research.
Hypothesis 1:
We expect to find clear evidence of negative campaign personalization. If parties are increasingly going negative in their advertising while at the same time leaders are becoming more central, we expect to see this negativity directed at competing party leaders.
Hypothesis 2:
The strategic nature of personalization suggests that not all leaders will be subject to the same degree of negative personalization. We hypothesize that negative personalization is related to the relative popularity of the party leaders. When a leader is unpopular, we expect competing parties to seize this opportunity by increasing their use of negative personalization. Conversely, when an opposing leader is popular, we expect competing parties to focus their attack on the party. Moreover, this may vary for the same leader over time as her popularity ebbs and flows—either between elections or within an election.
Case Selection and Data
To explore the possibility of negative personalization during election campaigns we examine two recent elections in Canada's largest province. Ontario serves as an appropriate case to explore our hypotheses for a number of reasons. First, the two elections under study occurred in close proximity to one another (2011 and 2014). This is important as longer periods between elections may result in high levels of campaign personnel turnover. In this case, however, many of the same individuals worked on both campaigns, which allows for a consideration of strategic decisions over time. Similarly, a leadership change within the Liberals prior to the 2014 election allows for an examination of whether opposition parties continued to rely on their previous strategies from 2011 or whether they adapted to the changing electoral environment. Finally, each of the major parties in Ontario can, to varying degrees, be described as a catch-all party, rejecting ideological rigidity and instead accommodating a diversity of opinions and groups (Cross et al., Reference Cross, Malloy, Small and Stephenson2015). This catch-all nature provides party leaders with considerable ideological flexibility and allows them to put their own stamp on the party.Footnote 1 This flexibility should offer a greater opportunity for personalization (both positive and negative).
One of the most common ways to examine personalization is through a content analysis of media stories (Kriesi, Reference Kriesi2012). Little work, however, has examined the content of the messages that parties themselves transmit. These party communications are the focus of our analysis. We examine party messages delivered in a number of different ways. First, 53 television advertisements were examined. These advertisements are particularly important because, as Cross (Reference Cross2004) notes, more voters will watch a campaign advertisement than will have any other interaction with the party. During the 2011 Ontario election political advertising was the single largest expenditure of the major parties representing more than half of their total expenditures (Elections Ontario, Reference Ontario2011).
Second, 368 party press releases that were issued during the campaign periods were examined. While party communications such as platforms and advertisements are largely static, press releases are dynamic forms of party messaging. They are issued on a near daily basis and therefore offer parties the ability to react to what is happening in the campaign and to change tactics from day to day. Although these documents are not necessarily produced for voters, they are attempts to frame media coverage of the campaign, and thus highlight issues and approaches that are strategically important to the party (Cross et al., Reference Cross, Malloy, Small and Stephenson2015; DiStaso, Reference DiStaso2012).
Although these are very different types of campaign communications, we are justified in including both in the same analysis for at least two reasons. First, diverging from the norm and including dynamic forms of communication (that is, press releases) provides a more inclusive picture of the campaign and how it unfolds (Hassell and Oeltjenbruns, Reference Hassell and Oeltjenbruns2015). Second, as Druckman and colleagues find, “candidates go negative with similar likelihoods” (Reference Druckman, Kifer and Parkin2010: 88) across a variety of media.
Each campaign communication was coded for whether it contained any mention of each of the main parties and their leaders. For television advertisements this includes not only verbal mentions but also images of the leader or the party logo. Excluded from the analysis are any information in the “paid for” disclaimer and any fleeting images of the leader or party logo.Footnote 2 Similar to the approach used by Dalton and colleagues (Reference Dalton, McAllister, Wattenberg, Dalton and P2000), coding campaign communications in this manner provides a ratio of leader-to-party mentions that can easily display the relative emphasis of each. The difference, of course, is that we are particularly interested in the leader-to-party ratio of negative campaign mentions by competing parties.
Negative and Positive Campaign Messaging: Evidence of Personalization
Considering both television advertisements and party press releases, Table 1 provides data illustrating the degree to which the focus is on parties and their leaders. Also included is the breakdown between positive and negative targets. Consistent with our first hypothesis, the data reveal clear evidence of personalized politics during election campaigns: parties not only frequently emphasize their own leader but also directly attack the leaders of opposing political parties.
Table 1 Campaign Communication by Target
Notes: Individual campaign communication may include mentions to both the party and leader. N refers to the total number of press releases or advertisements released during the campaign. Table includes 2011 and 2014 Ontario elections.
* = odds ratio p < 0.05 (comparison across rows).
Beginning with positive messaging (that is, mentions of your own party/leader), we find that both parties and their leaders are well represented in election campaign communications. During the 2011 and 2014 Ontario elections, for example, political parties mentioned their own party in more than 89 per cent of television advertisements and 80 per cent of press releases. Furthermore, party leaders were featured in 72 per cent of their party's television spots and 85 per cent of press releases. While party leaders do not always eclipse their party, the frequency of their mentions does highlight the importance of party leaders and their centrality during election campaigns. This is evidenced by a leader-to-party ratio of 0.81 for television advertisements and 1.1 for party press releases.
Evidence of negative personalization is also clear. As might be expected, television represents a powerful tool for negative personalization. Whereas opposing parties are mentioned in 15 per cent of television advertisements, rival party leaders are featured in more than half (53%). The press releases that parties issue on a daily basis contain considerable negative focus on party leaders as well, as two-thirds contain a reference to an opposing leader. Transforming these numbers into an opposing leader-to-opposing party ratio reveals a ratio of 3.5 for television advertisements and 1.2 for press releases. The 2011 election in Ontario, as detailed in the next section, is illustrative in this regard. During this election, the Progressive Conservatives and New Democrats (NDP) referenced the Liberal leader in 53 distinct campaign communications while only mentioning the Liberal party in ten.
The Strategic Decision to Personalize
Table 1 provides clear evidence of personalized politics in Ontario. However, it is also worth investigating the individual elections in order to uncover the campaign and party factors that influence a party's decision to personalize their opponents. Why do some parties focus their attack on leaders while others focus on political parties? Exploring more closely the 2011 and 2014 Ontario provincial elections, this section offers a preliminary exploration of the strategic factors influencing the decisions that parties make regarding the practice of negative personalization.
The 2011 Ontario Election
When the 2011 Ontario election was called, Dalton McGuinty had served as premier of the province for eight years and had been the leader of the Liberal party since 1996. During the lead-up to the 2011 election, however, there was a growing sense of McGuinty fatigue among Ontario voters (Cross et al., Reference Cross, Malloy, Small and Stephenson2015). While his government was relatively scandal free, the long tenure of McGuinty's leadership represented the status quo and engendered apathy with voters. The Conservatives and New Democrats, by contrast, both elected new leaders in 2009 and by doing so energized their parties and created a sense of renewal and excitement. By the time the 2011 election was called the electorate had enough time to form distinct opinions about each of the party leaders. A poll conducted at the beginning of the campaign revealed that McGuinty (Liberal), Hudak (PC), and Horwath (NDP) had vastly different approval ratings. By subtracting the difference of the approval and disapproval scores, we can create an approval index. Doing so reveals that three party leaders had approval ratings of −25, −10, and +12 respectively (Angus Reid, 2011; see appendix 1 for details on the approval ratings of party leaders).
While the Liberals did not completely abandon their leader during the campaign, they did recognize his declining popularity. In fact, the Liberals released a television advertisement at the beginning of the campaign in which McGuinty acknowledged his unpopularity (Cross et al., Reference Cross, Malloy, Small and Stephenson2015). Viewing McGuinty's rising unpopularity as an opportunity, the Progressive Conservatives developed a very clear strategy: forget the Liberal party and instead focus exclusively on the Liberal leader.
This strategy of negative personalization was consistent across multiple forms of party communication and endured throughout the campaign. Each of the eight television advertisements that the Conservatives produced for the 2011 election mentioned McGuinty, often referring to him as the “tax man.” By contrast, none of the eight advertisements made any mention of the Liberal party. This laser-like focus on the Liberal leader continued in PC press releases. Of the party's 34 press releases, 33 mentioned McGuinty at least once while only three mentioned the Liberal party (see Table 2). Even the Conservative election platform mentioned the Liberal leader nearly twice as often as it did the Liberal party.
Table 2 Advertisements and Press Releases by Target and Party (Ontario Reference Ontario2011)
Note: Individual campaign communication may include mentions to both the parties and leaders.
Although the Conservatives focused much of their attention on the Liberal leader, the party did on occasion target the NDP. When it did, however, the focus was universally on the New Democrats as a party and never on their relatively popular leader. This, of course, is in direct contrast to the approach that the PCs took with the Liberals where the emphasis was on the leader.
The Liberals also adopted different strategies based on the relative popularity of the opposing leaders. While the Liberals did not avoid targeting Horwath, their mentions of her never eclipsed mentions of the New Democrats. For example, the NDP and its leader were each featured in 57 of the 98 press releases that the Liberals produced. Similarly, both were mentioned in one of the fourteen television advertisements. By contrast, the Liberals attacked the more unpopular PC leader more than his party in both press releases and television advertisements.
For the most part the New Democrats ran a largely positive campaign, focusing on their own party and leader rather than targeting the Liberals or Conservatives. None of the NDP's television advertisements, for instance, mentioned any of the opposition parties or their leaders. When the NDP did target its opponents in press releases, it did so by focusing on the Liberal and Conservative leaders more than their respective parties. Overall the 2011 election supports our hypothesis concerning the strategic nature of negative personalization: negative personalization is less evident among popular leaders than it is among unpopular leaders. Horwath's popularity among voters ensured that she was never attacked more than her party and Hudak and McGuinty's relative unpopularity meant that they were always attacked more than their respective parties.
The 2014 Ontario Election
By 2014 the dynamics of the electoral landscape in Ontario had changed. Perhaps most importantly, the Liberals replaced Dalton McGuinty with Kathleen Wynne as the party's new leader. A poll conducted just after the election was called uncovered a number of similarities and differences between the political context in 2011 and 2014. Consistent with 2011, Horwath endured as the most popular leader, with an approval rating of −5. Unlike 2011, however, the PCs were now the weakest on leadership (−25) while the Liberals had a leader who was somewhat more popular with the electorate (−17) (Forum Research, Reference Research2014).
Given the increased relative popularity of the Liberal leader in 2014 (8 points on the approval index and no longer the least popular), we would expect the opposition parties to temper their efforts of negative personalization and focus more on the Liberal party and less on Wynne than they had McGuinty. As Table 3 illustrates, this is what occurred. While the Liberal leader was featured in every television advertisement that the Conservatives produced for 2011, this was the case in less than half of the PC advertising in 2014.
Table 3 Advertisements and Press Releases by Target and Party (Ontario 2014)
Note: Individual campaign communication may include mentions to both the parties and leaders.
A similar trend is evident for PC press releases as well. More than nine in ten featured the Liberal leader during the 2011 election compared to nearly 20 per cent fewer in 2014. The PCs also placed a greater emphasis on the Liberal brand than they had in the previous election. References to the Liberal party in PC press releases increased from 9 per cent in 2011 to 65 per cent in 2014. With an opposing leader who was no longer the least popular with the electorate, the PCs altered their strategy. Instead of focusing almost exclusively on the Liberal leader as they did in the earlier campaign, the party balanced its approach. This suggests that as the Liberal leader became more popular, negative personalization decreased and attacks on the party increased.
The Liberals turned their attention to the PC leader as he became increasingly unpopular with voters. While the 2011 Liberal campaign referenced the PC leader in 21 per cent of campaign advertisements, he was featured in all but one of the 2014 advertisements. Like the Conservative strategy used to target McGuinty in 2011, none of the 2014 Liberal advertisements made any mention of the Progressive Conservative party. In a television advertisement released near the end of the campaign, for instance, the narrator states that “On Thursday a split vote will only help Hudak's chances. Polls show only Kathleen Wynne can stop him now. Please consider voting for your local Liberal candidate.” The emphasis in this message is not about stopping the Progressive Conservatives but instead it is about stopping Hudak. Table 3 reveals that the Liberals attacked Hudak more than his party by a considerable margin at every opportunity.
Consistent with the 2011 election and our expectations, the NDP leader largely escaped being the target of negative personalization by both the Liberals and Conservatives. The PCs, for example, did not mention the NDP leader in any of their communications while they did reference the New Democratic Party in five press releases. Similarly, while the Liberals mentioned the NDP leader in 34 per cent of their press releases, the party was featured in 41 per cent.
It is worth illustrating the strategy of negative personalization with some examples from the campaign. The following excerpts from three press releases perfectly capture the strategic element of negative personalization. In each case, the party issuing the press release sought to personalize one opponent (the weaker party leader) while attacking the other party's brand (the stronger party leader).
The Hudak PCs’ regressive, ideological opposition to a strong public postsecondary education sector would take Ontario down a path toward a low-wage, low-growth economy. Both Tim Hudak and NDP will put our success at risk. (Liberal party press release, May 7, 2014)
Neither the McGuinty-Wynne Liberals nor the NDP have a plan or the discipline necessary to focus on jobs every single day. (Progressive Conservative party press release, May 7, 2014)
This election Ontarians have a choice, they don't have to endorse a corrupt Liberal party or Tim Hudak's nonsensical plan to add 100,000 people to the unemployment rolls. (New Democratic Party press release, May 22, 2014)
These examples very clearly highlight the strategic nature of negative personalization. Within a multi-party electoral arena, parties make strategic decisions about which opposing leaders to negatively personalize and which party brands to attack. Rather than embracing negative personalization as a blanket strategy, parties use it as a tool to further undermine already unpopular leaders. The Ontario example also provides preliminary evidence of strategic decisions over time. With the change of leaders and their popularity, the targets of negativity changed as well. That is, the parties did not rely on the same strategies in 2014 as they did in 2011. Instead, they adapted to the changing electoral environment and responded to changes in leadership popularity and strength.
We can also explore the time dimension within an election as well as between elections. In this regard, the 2014 Liberals are a good case to examine as Wynne's popularity among voters fluctuated over the course of the campaign.Footnote 3 Figure 1 plots the Liberal leader's popularity (approval score) at five different points over the course of the 2014 election campaign (see appendix 2). Also included in this figure is the level of negative personalization found in PC press releases targeting the Liberal leader during the campaign at the same intervals. Issued almost daily, the dynamic nature of press releases allows parties the ability to change their strategy quickly during the campaign and respond to public opinion. In this case, we find that the PCs targeting of Wynne was relatively sophisticated in that it responded to shifts in public opinion throughout the campaign. The Progressive Conservatives negative emphasis on Wynne is the lowest of the entire campaign period when she is at the height of her popularity in week three. Conversely, attacks on Wynne compared to her party are the highest when her popularity plummeted in week four. Figure 1 therefore provides further evidence to support our hypothesis regarding the relationship between negative personalization and leader popularity.Footnote 4
Figure 1 Negative Personalization and Leader Approval (Liberal, Ontario 2014 Election)
While the Liberals issued considerably fewer press releases (29, compared to the 107 issued by the PCs), we can replicate the above figure for the treatment of the PCs by the Liberals. Unlike Wynne, Figure 2 demonstrates that Hudak's (un)popularity was relatively stable throughout the campaign period. Interestingly, we find that the Liberals use of negative personalization was also stable throughout the four weeks that they issued press releases. Consistent with our expectations, this suggests a strategic use of negative personalization.
Figure 2 Negative Personalization and Leader Approval (PC, Ontario 2014 Election)
Conclusions
In this research note we aimed to make a number of theoretical and empirical contributions to the growing literatures on personalization and negative campaigning. The central contribution is an examination of the inter-party dynamics of campaign personalization and the development of a new concept: negative personalization. In adopting this approach we questioned whether parties could play a role in personalization by negatively personalizing their opponents. In particular, we hypothesized that negative campaign personalization would be a common feature of election campaigns and that this negativity would be targeted at unpopular leaders more so than popular ones.
Consistent with our first hypothesis, we provide compelling, albeit preliminary, evidence to demonstrate that negative personalization is a common feature of contemporary election campaigns. That is, parties routinely attack opposing party leaders in addition to the parties that they lead. In fact, our analysis of the 2011 and 2014 provincial elections in Ontario demonstrate that both television advertisements and press releases are significantly more likely to mention an opposing party leader than an opposing party. This is particularly evident in television advertising where more than half of all campaign ads targeted an opposing party leader compared to only 15 per cent that mentioned an opposing party. The predominance of negative personalization in television advertising is consistent with the broader personalization literature, which has noted the ease with which leaders, or opposing leaders in our case, can be pictured on screen and the impact that these images can have on voter perceptions (see Karvonen, Reference Karvonen2010).
Considering our second hypothesis, we find clear evidence that negative personalization is indeed a calculated decision. Parties make strategic use of their campaign messaging and in this regard attack their opponents where they are the weakest: party leaders who are popular experience the least negative personalization while relatively unpopular leaders experience the most. This helps to explain why the New Democrat leader escaped negative personalization in 2011 and 2014 while the PC leader did not. It also explains why the Liberal leader was subject to more negative personalization in 2011 than his more popular successor in 2014. Furthermore, an examination of the dynamics of negative personalization over the course of the 2014 Ontario election campaign reveals that the Progressive Conservative strategy responded to shifting public opinion and targeted the Liberal leader the most when her approval was lowest.
While outside the scope of this study, the analysis presented in this research note raises a number of questions that require further attention. First, campaigns are dynamic events and the decision to negatively personalize an opponent may change over the course of the election. While we explore some of these campaign dynamics in terms of popularity, more can be done. We know from the negative campaigning literature, for example, that campaigns tend to begin positive and become more negative as election day nears (Damore, Reference Damore2002). Future research may consider whether similar trends define the practice of negative personalization.
Second, popularity of the leader is not the only strategic consideration that parties face. Other considerations, such as the competitiveness of the election, may play a role as well. Similarly, parties may adopt different strategies depending on the nature of their competitors. Take, for example, leader-centric parties. When competing against a leader-centric party, an attack on the party leader is likely the most effective attack on the overall party brand.
Finally, while negative personalization is a strategic decision, party system factors likely shape the options available to parties. Parties competing in ideologically fixed and ideologically flexible party systems, for instance, may behave very differently with regards to negative personalization. Longstanding and entrenched ideological differences naturally emphasize parties, brands and issues over leaders and personalities. As such, we might expect personalization to be suppressed in ideological party systems, as campaigning will be focused on parties and issues. By contrast, ideologically flexible party systems offer a greater opportunity for personalization (both positive and negative).Footnote 5 However, these propositions, as well as the hypotheses examined in this note, should be tested in other cases before firm conclusions can be made.
Appendix 1: Party leader approval ratings
Appendix 2: Party leader approval ratings (during 2014 election)
