Candidate campaigning in parliamentary systems

Abstract

The individualization of politics is usually studied in relation to party leaders. Using new data from the Norwegian Candidate Survey 2009 and in-depth interviews with 29 top candidates, in this article we study whether candidates in the Norwegian 2009 parliamentary election ran party-centred or individualized campaigns. We distinguish between the organizational aspects and the communicative focus of the candidates’ campaigns. Moreover, we argue theoretically and show empirically how campaigns can be localized but still party-centred. The analysis shows that there are low levels of individualized campaigning in Norway, but that there are differences between candidates, especially based on party affiliation. Moreover, the differences are first and foremost related to the communicative focus, not the organizational aspects. While candidates highlight the importance of localizing the campaign, the results show that this is mostly about ‘translating’ the national campaign strategy to the regional or local level, not about independent local strategies.

Keywords

Campaigns campaigning candidate campaigning individualized campaigning party-centred

Introduction

The waning of cleavages and the alleged ‘decline of parties’ has increased attention on the role of individuals in politics (Dalton and Wattenberg, 2000; Karvonen, 2009; McAllister, 2007). Developments in the media and the dominant media logic in political campaign coverage are also said to increase this tendency (Swanson and Mancini, 1996). Indeed, at the electoral level, some argue for a shift from party control to domination by the leaders (Poguntke and Webb, 2005: 10). This increasing focus on and importance of individuals is often labelled the personalization of politics (Karvonen, 2009; Kriesi, 2012; McAllister, 2007). The role of individuals has attracted attention particularly in studies of election campaigns mostly focusing on the increasing importance of party leaders (Kriesi, 2012).¹ However, election campaigns are fought not just by party leaders, but also by the candidates standing for election. A focus on individual candidates represents what Balmas et al. (forthcoming) label decentralizing personalization. In the United States, the tendency towards individualization is evident among candidates running for Congress to the extent that the system is labelled ‘candidate-centred’ (Agranof, 1972; Brox and Shaw, 2006; Plasser and Plasser, 2000).² Candidates build their own campaign organizations, and the candidates themselves more than their parties are the focus of the campaign communication. Less is known about candidate campaigning in Western Europe. In this article, we investigate whether individualization and personalization are features of candidates running for parliament, and ask if they run party-centred or individualized campaigns.

Using data from the Norwegian Candidate Survey 2009 and in-depth interviews with 29 top candidates in the seven parties represented in parliament, we study individual candidates running for election at the Norwegian 2009 parliamentary election. We distinguish between organizational aspects and the communicative focus of the candidates’ campaigns. More specifically, we ask to what extent and in what sense the candidates are part of a greater party campaign organization and comply with the centrally decided campaign strategy. We ask what their communicative focus is, i.e. the party or their own candidacy. Earlier studies have barely distinguished between individualized and localized campaigning. In this article, we argue theoretically and show empirically how campaigns can be localized but still be party-centred.

We aim to explain the role of candidates by developing a theoretical framework addressing factors that influence candidates’ campaign behaviour. The rise of the candidate-centred campaign was already evident in the US in the late 1960s (Agranof, 1972). In Western Europe, on the other hand, countervailing forces might curb tendencies towards individualization at the candidate level. In this sense Norway is an interesting case. The theoretical discussion below shows that key characteristics of the Norwegian campaign environment point in the direction of low levels of individualized campaigning. However, although our results indicate that the Norwegian campaign is indeed party-centred, there are differences between candidates, especially based on party affiliation.

The article proceeds as follows: We begin by addressing the centralization thesis as found in the party and campaign literature (e.g. Farrell and Webb, 2000; Katz and Mair, 1995; Norris, 2000). We then discuss candidate campaigning and distinguish theoretically between the organizational/strategy aspects and the communicative aspects, and argue for a distinction between individualized campaigning and localized campaigning. In the next section we discuss the factors that might influence the nature of candidate campaigning and discuss what to expect from the Norwegian case. In the following empirical analysis, the candidate survey and the interviews are used to study the campaign behaviour of the candidates. First we investigate the organizational and strategic aspects and second the communicative aspect. In the third part of the analysis we relate the two aspects of individualized campaigning to localized campaigning. In the final section we discuss the empirical findings in relation to the framework sketched out initially.

The vertical election campaign structure

Increasing centralization, meaning more control of the campaign strategy at the national level, is pivotal in what we can label the modernization perspective of campaign change. Several scholars have argued for a new phase of campaigning which is characterized by, among several other factors, high levels of centralization (Farrell, 1996; Norris, 2000; Farrell and Webb, 2000). It is argued that parties combine a strong focus on the district with a highly centralized party-driven coordination (Norris, 2000). In these studies, centralization tendencies in campaigns are often understood as part of a general centralization process in political parties (cf. Katz and Mair, 1995). However, the vertical structure of campaigns differs between systems. Plasser (2002) distinguishes between a US style and a West-European style of campaigning. To increase democratic input, American political parties in the late 1960s changed their nomination process and introduced the primaries. One result of the reforms was that candidates set up campaign organizations and expertise independently of the party (Agranof, 1972; Brox and Shaw, 2006). Consequently, as mentioned initially, the US style is candidate-centred, meaning that the campaign organization is built up by the individual candidate almost from scratch. Candidates hire a campaign manager and additional campaign staff. They identify their own campaign strategy, conduct their own polling and organize volunteers in grassroots campaigns (Herrnson, 2007). West-European campaigning is labelled party-centred and the centralizing efforts of the parties are emphasized in this model (Plasser, 2002).³

Candidate campaigning and the local aspect

As mentioned initially, in this article we distinguish analytically between two aspects of candidate campaigning:⁴ (i) the organizational and strategic dimension ⁵ and (ii) the communicative dimension.⁶ The organizational and strategic dimension is the relationship between the central party and the candidates. One way of thinking about this relationship is as a dimension where, on the one extreme, the candidates build their own organization with few or no organizational links to the central party leadership of whichever platform they are running. On the other, the party is the campaign organization and the candidates an integrated part of this collective: Candidates depend on the party for organizational resources and they comply with the centrally determined campaign strategy. The communicative focus of the campaign refers to the focal point of the candidates’ communicative aim. On the one hand, the communicative goal of the candidates might be to create as much attention as possible for themselves. On the other hand, it might be to draw as much attention as possible to their parties. These two dimensions are combined in Figure 1.

Figure 1.

Two dimensions of candidate campaigning: Organization and communicative focus.

The concept individualized campaigning refers to a situation where the candidates campaign independently of the party (Zittel and Gschwend, 2008: 980). In the figure, the diagonal from the bottom left to the top right corner can be considered the main individualized campaigning dimension. The lower left area of the figure represents low levels of individualized campaigning, whereas the upper right area represents high levels of individualized campaigning. The two aspects might be expected to correlate with each other. A candidate who builds her own organization independently of the party will most likely also focus more on her own candidacy. However, this is an empirical matter, and candidates who are part of a greater party campaign organization might still focus on their own candidacy in the campaign.

Earlier studies have seldom differentiated between individualized and localized campaigning. We argue that a campaign might be localized in both organizational and communicative terms, but still be party-centred on both dimensions. In other words, candidates might run highly party-centred and localized campaigns. Localizing is perhaps most relevant for the ‘organization and strategy’ aspect of campaigning. Local party branches might organize their campaign and determine their own strategy independently of the central party without it being centred on one candidate. However, campaigning can also be localized and party-centred in regard to the communicative focus: The ‘local party’ as opposed to the ‘national party’ can be the focal point of the communication. For example, in highly decentralized systems local party branches might want to differentiate the ‘local party’ from the central party. The distinction between individualized and localized campaigning is scrutinized in our empirical analysis.

Factors that influence the role of candidates in the vertical campaign structure

The US and West-European styles of campaigning should be considered ideal types, especially as there are differences from one West-European country to the next concerning styles of campaigning. The extent of the difference in regard to individualized campaigning is an empirical question, and we argue that it will vary according to constitutional type, electoral systems and, most importantly, the nature of the nomination process.

It is often claimed that presidential systems increase personalization because the focus is on the candidates running for the presidency. Hence, at the national level one can expect presidential systems to be more personalized than parliamentary systems (Kriesi, 2012: 827). It is less apparent that presidential systems will increase the individualization among parliamentary candidates. However, top-down control of the party organization is more important in parliamentary than in presidential systems. In parliamentary systems the executive power and an effective government depend on party discipline in Parliament. Hence, in parliamentary systems the party leadership must maintain party discipline to secure party government, and are probably less inclined to implement reforms that will weaken their control of the party organization.

Several scholars argue that proportional systems will increase party-centred campaigning (Bowler and Farrell, 1992: 8; Plasser, 2002; Swanson and Mancini, 1996: 17 f.). According to Bowler and Farrell (1992: 8), proportional representation list systems promote greater centralization than plurality systems, and Swanson and Mancini (1996: 17) claim that majority systems highlight the role of the individual candidate and thus promote what they call personalization. However, Plasser (2002) maintains that the empirical evidence points in the direction of party-centred campaigning in the West-European systems, with plurality systems, most importantly, in Britain and France. The empirical evidence from Britain appears mixed, however: Denver et al. (2003) support the centralization perspective. They claim that constituency campaigning was reinvented in the 1990s and that the central party has become increasingly important in planning and managing constituency campaigns. Recently, Fisher and Denver (2008: 821) argued that there was a move away from traditional grassroots activism towards modern and centrally directed campaigning in Britain. Whiteley and Seyd (2003), in a study of the British Labour 1997 campaign, argued on the other hand that the variations in campaigning between the constituencies were produced by the candidates (and local branches) rather than by the targeting efforts of the central party. However, despite their differences, these studies show that UK constituency campaigning is clearly different from the candidate-centred US campaign. Consequently, although plurality systems provide incentives towards individual campaigning, the electoral system cannot be the decisive factor.

The organizational weakness of the US parties is considered a major explanation for the candidate-centred US campaign (Agranof, 1972), a weakness increased by the introduction of the primaries (Brox and Shaw, 2006).⁷ Consequently, control of the nomination processes appears to be a key factor (cf. Karlsen and Narud, forthcoming). Two dimensions of candidate selection are given special attention in the nomination literature (e.g. Gallagher and Marsh, 1988). One is the degree of centralization, i.e. what part of the party supervises the selection of candidates. The second is the degree of inclusiveness in the process, i.e. who is allowed to participate in the selection? If we combine these aspects the control of nominations may vary between, on the one extreme, the party leader or central party leadership and, on the other, the electorate deciding the nomination. For example, if voters decide, candidates must highlight their own candidacy in competition with others from the same party platform; if the party decides, candidates must cater to the party.

The effects of the nomination process must be considered in relation to the electoral system (Karlsen and Narud, forthcoming). Zittel and Gschwend (2008), in their study of candidate campaigning in the German mixed-member electoral system, show that candidates who ran as direct candidates were modestly more individualized in their campaigning efforts. A different and perhaps stronger incentive for individualized campaigning is found in proportional systems with preference voting schemes: open lists where the voters decide the final ranking of the candidates (cf. Carey and Shugart, 1995; Grofman 2005). In such systems, candidates must promote themselves and compete with other candidates from their own party, thereby causing intra-party competition (Plasser and Plasser, 2002).

Based on this discussion, we expect campaigning in Norway to be party-centred and highly centralized, with low levels of individualized campaigning in regard to both the organization of the campaign as well as the candidate’s communicative focus.⁸ There are three main reasons for this expectation.

The Norwegian political system is characterized by a parliamentary government, a stable multiparty system and well-organized membership parties. Therefore, first, as discussed above, top-down control and party discipline are essential in a parliamentary system like the Norwegian system in ensuring effective government. Second, the parties control the nomination process, which is decentralized, as the nominations are made by representative conventions organized by the constituency branches of the party organization (Valen et al., 2002). As argued above, control of the nomination strengthens the role of the parties. Third, the electoral system is based on proportional representation and a modified Sainte-Laguë system has been employed since 1952. There is only a theoretical possibility of influencing the candidate order.⁹ When the candidate is nominated, his or her election depends solely on the party vote. Consequently, there are few reasons for candidates to cultivate a personal vote. The question is to what extent such an environment still allows for individualized campaigning.

Variation between candidates

Although the campaign environment will have an effect on the vertical structure of campaigns and the overall level of individualized campaigning, the role of candidates and their behaviour might still differ among candidates in the same system. The electoral system is important in this regard as well. In a proportional system a number of candidates from one party might expect to win a mandate, while a number of lower placed candidates will stand no chance of being elected. The ones with no chance of obtaining a mandate will probably campaign to a lesser extent than higher placed candidates, and act more like party activists in plurality systems. However, it is not easy to predict whether these top candidates should be more individualized or more party centred in organizational and communicative terms. In regard to organizational focus, they might be more inclined to run their own campaign; however, they also need resources, and the central parties might largely control these resources. In regard to communicative focus, the top candidates might be more inclined to focus on themselves. However, too much focus on own candidacy might not go down well in the party. Consequently, candidate behaviour might also differ based on party affiliation, as the tolerance for individual campaigning most likely differs from party to party. Such differences may originate in variations in the ideological and cultural background of the parties. For example, the Norwegian political system consists of three levels of government: local, regional and national. All major parties have adopted an organizational structure that corresponds to this structure (Heidar and Saglie, 2003), and the top-down governed mass party (Duverger, 1954/1972) has traditionally been the preferred organizational model for Norwegian parties (Heidar and Saglie, 2003). In a study of policy influence in the Norwegian party organizations, Allern and Saglie (2012) found that there are some differences in the acceptance of top-down culture among the parties. The Conservatives, the Centre Party, the Christian Democratic Party and the Socialist Left Party claimed that top-down control was alien to their party culture. Top-down culture was more accepted in the Labour Party and the Progress Party. Top-down acceptance in day-to-day party work does not necessarily transfer to intensive campaigning and the relationship between the central campaign headquarters and individual candidates. However, these patterns can be related to the parties’ historical background: The Labour Party emerged outside parliament and the importance of the party organization as well as party discipline has been emphasized, whereas the Conservatives emerged within parliament and the emphasis has to a greater extent been on the parliamentary group. The mass party ideal is evidently stronger and more accepted in the Labour Party and the Progress Party, and we expect candidates from these parties to be more reliant on the party in their campaigning efforts and the candidates’ communicative focus to be more on the parties than on their own candidacy.

Data and method

We study candidate campaigning in Norway with the help of two data sources; first, the 2009 Norwegian Candidate Survey (NCS). The NCS is a survey of all candidates running for election for any of the seven major parties in the 2009 parliamentary election.¹⁰ The survey was distributed as a post-enquete.¹¹ The original sample consisted of 1972 candidates and the response rate was 52 percent, leaving us with 1015 candidates included in the study. All parties are almost equally represented among the candidates included, and top candidates were equally represented to the lower placed candidates. Second, we utilize in-depth interviews with 29 candidates in four constituencies. The interviews were semi-structured and each lasted approximately one hour. Thirteen of the top candidates and 16 second-place candidates were interviewed. All parties and constituencies were about equally represented in the material: Four to five candidates from each party and six to seven candidates from each constituency.

In the survey, we operationalize candidates’ campaign organization and role of the central parties with the question: On a scale from 1 to 5, where 1 indicates not important and 5 very important, how important was the central party and the work being done there for you in your campaign efforts? The communicative focus of the campaign is measured using the following questions: What was the primary aim of your campaign? Where would you place yourself on a scale from 0 to 10, where 0 means ‘to attract as much attention as possible for me as a candidate’ and 10 means ‘to attract as much attention as possible for my party’? The localization of campaigning is measured with the following question: Did you raise any issues during your campaign that were specific to your constituency and that were not raised by the national or regional party?

Above, we argued that candidates’ behaviour might differ based on their list placement, and that this is related to the expectation of winning a mandate. Candidates with a chance of winning a mandate can be expected to campaign to a greater extent. In the analysis we mostly use the list position to study such differences, and not their expectations of obtaining a mandate.¹² The number of candidates expecting to win seats will differ between parties based on their size, but we nevertheless believe it is the best strategy to compare the top candidates in the different parties, as they can be expected to ‘lead’ the campaign in their constituency.¹³

In the analysis we report whether the main differences are statistically significant. However, our data are not based on a sample of the population; the whole population is included in the sample and consequently there is no need to test if the differences are statistically significant for the ‘population’. As statistical significance to a great extent depends on the number of units in the groups compared, and there are few top candidates in each party, it is hard to achieve statistical significance. Therefore it is essential to keep in mind that the differences we report are the real differences between the candidates who ran for election at the 2009 election.

In the following analysis, we use the combination of the quantitative and qualitative data to describe, discuss and explain similarities and differences between candidates. Instead of separating the different types of data in the analysis of the candidates, they are juxtaposed to highlight, contrast and illustrate the points.

Empirical analysis

Theoretically, we distinguished between the organization/strategy aspect and the communicative focus of candidate campaigning. Moreover, we argued that localized campaigning entails something different from individualized candidate campaigning. Consequently, the empirical analysis has three main parts. In the first, we investigate the organization/strategy aspect of candidate campaigning, and in the second the communicative focus of the candidates’ campaigns. In the third part we explore the extent to which the candidates’ campaigns are localized and relate this to the two different aspects of individualized campaigning.

The vertical organization and strategy

To what extent and in what sense are the candidates part of a greater party campaign organization? From the theoretical discussion we expected candidate campaigning to be predominantly party centred in organizational terms. The parties are strong, they control the nomination process, and there is no opportunity for preference voting. Hence, we argued that the campaign environment facilitates a strong central party campaign organization. Sixty-five percent of the candidates indicated that the central party was important for them by assigning themselves 4 or 5 on a scale from 1 to 5 (5 = party very important), and the mean for all candidates was 3.7. This is a first indication of the importance of the central party. However, as discussed above, we also argued that we were likely to find differences between candidates based on placement on the list, or the likelihood of being elected. Figure 2 reports how the candidates assessed the role of the central party after their placement on the party lists.

Figure 2.

The importance attributed to the central party for the candidates’ campaigning on a scale from 1 to 5, after placement on the list.

The figure shows that the top candidates value the work done by the central party to an even greater extent than lower placed candidates. As argued, the top candidates stand a good chance of being elected and they campaign more extensively than lower placed candidates. Consequently, they are more likely to be in need of resources and assistance from the central party. The in-depth interviews support this argument. No. 1 candidates expressed adherence and knowledge of the central party strategy, whereas No. 2 candidates, who had little or no chance of being elected, expressed that they were less integrated in and less bound by central strategies. No. 2 candidates with a secure mandate or with specific central positions in the party, on the other hand, put similar weight on the central party’s campaign strategy as the No. 1 candidates. The candidates who considered it possible to get elected seemed to campaign more wholeheartedly and consequently relied more on the party.

None of the in-depth interviewed candidates regarded the central campaign strategy as unimportant or uninfluential. They all saw the central campaign as the framework for their campaigns. However, the coordination between the candiates and the central level varied among candidates, as did the way they used the local political context to front local issues. We return to this aspect below when we study the localizing efforts of the candidates.

Above, we discussed how party affiliation might have a bearing on how candidates relate to the central party campaign strategy, as the culture and tolerance for individualized behaviour most likely differ between parties. The survey points to some minor differences between candidates based on party affiliation. Figure 3 presents the importance attributed to the central party for all candidates and the No. 1 candidates in the seven parties.

Figure 3.

The importance attributed to the central party for the candidates’ campaigning on a scale from 1 to 5 for all candidates and the No. 1 candidates in the different parties (standard deviation).

On average, candidates in all parties are well over on the ‘party important’ side of the scale. As mentioned, no candidates, regardless of party, viewed the central campaign strategy as unimportant or uninfluential in the in-depth interviews. The candidates from the Centre Party (Sp) are most party-oriented, whereas the Progress Party have the fewest centrally oriented candidates. As indicated by Figure 1, in all parties the No. 1 candidates were more inclined to consider the central party important than lower placed candidates were. The exception is the Progress Party. In the in-depth interviews the candidates from the Progress Party (FrP) painted a picture of rather loosely organized local campaigns that were more or less independent of the central, high profile campaign, thereby supporting the survey findings. However, it is not about an individualized campaign set-up, it is rather that they did not receive the assistance they wanted from the central party. Unlike the Labour and Centre Parties, where there was daily coordination between the central party organization and the regional campaigners, this seemed not to have been the case in FrP. The central party leadership ran a nationwide, high-cost campaign with only marginal coordination with the local campaigns to the degree that sometimes the candidates were surprised by actions or initiatives taken by the party leadership:

The communication between the central and the local level could have been better. We did not always know when the Party Leader, Siv Jensen, launched an issue or was quoted in the media. Sometimes we risked being called by a journalist telling us that ‘Siv Jensen said this, what is your opinion on this issue?’ (Interview 8/10/09)

To conclude this part, candidates – especially the higher placed ones – consider the work done at the central party pivotal for their own campaigning.¹⁴ Candidates are integrated in the greater party campaign organization and comply with the party campaign strategy. In regard to the organizational and strategic aspects of candidate campaigning, campaigning in Norway is clearly party centred.

Figure 4.

The communicative focus of the campaign after placement on the list. Mean on a scale from 0 to 10 (standard deviation).

Communicative focus

In the theoretical discussion above, we expected candidates in Norway to focus on the party in their campaign communication. As mentioned, parties are strong, they control nomination, and there is no possibility of influencing the candidate list order through preference votes. To a great extent, the expectation of party centred campaign focus is met. Fifty-eight percent of the candidates placed themselves at the extreme 10 (10 = attract as much attention as possible for the party), and 80 percent placed themselves at 8, 9 or 10. The average mean for all candidates is 8.7. However, there are some interesting differences based on placement on the list, as shown in Figure 5.

Figure 5.

The communicative focus of the campaign. Mean on a scale from 0 to 10. All candidates and the No. 1 candidates after party affiliation.

The inclination of candidates to focus on themselves appears to be a function of their position on the list. Above, we saw that the top candidates were more inclined than lower placed candidates to say that the central party and the work done there was important to them in their own campaigning efforts. Nevertheless, the No. 1 candidates are less inclined to campaign to attract as much attention as possible to the party, thereby placing themselves towards the centre of the scale, whereas candidates lower on the lists prioritize the party. Hence, there are different tendencies for the two aspects of individualized campaigning we distinguished between.

The in-depth interviews also shed light on this phenomenon. In general, the candidates that did not have a secure mandate were more or less party workers. Since they had little or no theoretical chance of being elected, their campaign efforts mainly consisted of promoting the party, resembling party activists in majority systems.

The communicative focus of the campaign does not just vary among candidates based on list placement. We also expected this to vary between candidates from different parties. As Figure 5 reveals, there are distinct differences based on party affiliation.

As expected from the discussion of the Labour Party (Ap) and the Progress Party (FrP) as being top-down controlled, the No. 1 candiates from these parties are least inclined to focus on their own candidacies. The Labour candidates were clearly the most party-oriented in their communicative focus, followed by the Progress Party candidates, whereas the Christian Democratic Party (KrF) and the Conservatives stood out as the most individual-oriented. The interviews add evidence to this picture. When asked about their campaigning strategies, the candidates from KrF, the Conservatives (H) and the Socialist Party (SV) as a rule talked more about individual factors than the Labour candidates. This quote from a KrF candidate is a good example:

I created the media and communication strategy for (county/constituency) myself. During the course of the campaign I gave a personal interview to (the main regional newspaper) showing my private side as a person who enjoys the good things of life. After that I received a lot of criticism from the ‘core voters’ of the party, and as a result I contacted the central party organization for advice. My situation was discussed in the central party and I received 150% support for being who I was. In KrF there is a vast distance between real politics and the public image, and I wanted to change that. (Interview 5/10/09)

Interestingly, although the candidate focused on herself, she still wanted to get support from the central party when she was critized by local party activists. A candidate from the Socialist Party replied when asked whether it was him or the party that outlined the campaign strategy in his constituency:

Now, I also want to do my own things. I sometimes choose to do things that are not fully in line with the party program, taking the responsibility for and control of that myself. Concerning transportation and communication issues, which are my main responsibilities in Parliament, I maintain the party line. But on foreign policy issues I sometimes diverge. (Interview 22/9/09)

Both these candidates emphasized that they wanted to draw attention to themselves as well as to the party. Interestingly, both also argued that their diverging from the central party line or party image was beneficial to the party, as it signalled room and space for candidates with diverse views, thereby indicating that their party cultures allowed for individual campaigning strategies.

A typical example from a Labour candidate rather focuses on the importance of the party in the localized campaign strategy:

For the Labour Party it was important to show what we had obtained in my constituency. My experience was that we had succeeded in establishing enough kindergartens; the municipalities’ economies were considerably improved; we had increased investments in transport and communication, etc. (…) Labour won votes all over my constituency which I see as evidence that our message was understood by the voters. (Interview 13/11/09)

Figure 6.

Localized constituency campaigning. Proportion of candidates who campaigned on an issue of importance to the constituency which was not on the central party agenda after placement on the list.

The inclination of top candidates to focus on themselves might be due to factors other than list placement and party affiliation. Top candidates are typically men and they are older than lower placed candidates. Studies suggest that male politicians are more publicity-seeking than women (Aalberg and Strömbäck, 2011). We have less clear-cut expectations based on earlier research in regard to the effect of age. However, we might expect older candidates to be more secure in their roles and focus more on their candidacy than younger candidates do. To control for such factors, in the final analysis we study the effect of age, gender, list placement and party affiliation on candidate communicative focus. We also investigate whether the organizational aspect – operationalized with the question on the importance of the central party – affects communicative focus. We have already seen that the effect of party was partly contingent on list placement. Hence, to deal with this interaction effect, in Table 1 we report the results of two multivariate analyses: one for candidates below the top 3 and one for the top 3 candidates in all parties.

Table 1.

The effects of age, gender, list placement and party affiliation on individual communicative focus. Separate analyses for candidates placed below top 3 and the top 3 candidates Multivariate regression analysis.

	Below top 3 (N=706)		Top 3 (N=199)
	Coeff.	S.E.	Coeff.	S.E
Constant	2.34**	0.38	4.34**	1.18
Age	–0.01	0.01	0.001	0.02
Gender (female)	–0.21	0.16	−0.96**	0.35
List placement	–0.05	0.02	−0.68**	0.21
Party important	–0.07	0.06	−0.23	0.18
Party affiliation
SV	–0.37	0.27	0.25	0.62
Sp	–0.74*	0.29	1.12	0.61
KrF	–0.46	0.28	0.37	0.61
V	–0.22	0.29	−0.16	0.65
H	–0.15	0.30	1.78**	0.60
FrP	–0.08	0.28	0.31	0.61
R²	0.01		0.15

**p < 0.01; *p < 0.05.

Dependent variable (0–10, 0=party focus, 10=individual focus).

Age is continuous from 0 (recoded from 18).

Gender: male= 0 female = 1.

List placement is from first (1) to last place.

Party affiliation: reference category is Labour Party affiliation.

In the model for the candidates below the top 3, none of the variables has any effect apart from Centre Party affiliation. The main reason for this is that there is very little variation in the dependent variable, and consequently there is little to be explained. In the model for the candidates in the top 3, gender, list placement and party affiliation all have effects. Women are less likely to focus on their own candidacy than male candidates are, and this fits well with earlier research finding men to be more publicity-seeking than women (Aalberg and Strömbäck, 2011). Higher placed candidates are more likely to focus on their own candidacy even when gender and age are controlled for. Moreover, the conservative candidates are more likely to focus on themselves even when age, gender and list placement are controlled for. As could be expected from the analysis above, which showed that higher placed candidates were more likely to find the party important for campaigning, there is no effect of independent campaigning in organizational terms on individual communicative focus.¹⁵

Figure 7.

Proportion of candidates that campaigned on local issues after communicative focus.

In terms of communicative focus, as expected based on the theoretical discussion, candidate campaigning in Norway is predominantly party focused. However, top candidates in some parties are more inclined to focus on their own candidacy. As discussed above, we relate this to differences in party cultures when it comes to the acceptance of individualized focus. We return to this discussion in the final section.

Localized campaigning

Above, we argued that conceptually, individualized and localized campaigning should be distinguished one from the other; campaigns can be both localized and party centred. Our data suggest that candidate campaigning in Norway is highly localized. The level of localized constitutency campaigning among the party candidates is quite high: Approximately half of the candidates raised a local issue.

The tendency to focus on local issues is even more pronounced among the candidates who are placed high on the list. Seventy percent of the top candidates raised a local issue, but even 37 percent of the candidates who were placed towards the bottom of the list campaigned on local issues. The material reveals minor differences between the parties in this regard. The tendency to focus on local issues is widespread in all parties. Nevertheless, it is evident from the in-depth interviews that the localizing strategy is mostly about making the national campaign relevant at the regional and local level, not about developing independent local strategies. Some candidates focused on presenting the central party campaign issues in localized terms, as this candidate from the Progressive Party stated:

There are few differences between central and local issues. They are basically the same. In my constituency it is transport and communication, health and education. When I tour my constituency and talk about these issues; I use local examples that are important for my own voters. At the central level they talk more generally. (Interview 8/10/09)

Other candidates focused on specifically local issues, as did this Conservative top candidate:

I was mayor for a long period and had many contacts among journalists. We had profiled ourselves on important issues well before the election campaign started. Among the issues I worked with in my constituency was a high-voltage electricity line, a road project between two central municipalities, and an issue concerning taxation of a local industry. All of these produced several media items, and focused on issues important for the region. (Interview 2/12/09)

So, candidate campaigning is localized, but, as these examples show, campaigns were localized within the centrally decided party campaign framework. Let us take a closer look at the relationship between individualized and localized campaigning. A correlation between the two aspects of individualized campaigning and localized campaigning shows that there is no relationship with the organizational aspect, but a significant positive, but not too strong, relationship between individual focus and localized campaigning (0.16). However, if we distinguish the candidates who can be said to have an individualized focus from the ones who focus on the party,¹⁶ we find that although candidates with individualized focus are more prone to focus on local issues, it is evident that both types of candidate campaign on local issues to a great extent. This supports the argument that localized campaigning must be distinguished from both organizational and communicative individualized campaigning.

Discussion and conclusion

Initially, we discussed and identified three factors which are likely to influence the nature of candidate campaigning. Norway is a parliamentary system, the parties (locally/regionally) control the nomination of candidates, and the electoral system is based on a proportional system with no preference voting scheme. We argued that these system traits strengthen the role of parties and offer few incentives for the candidates to run individualized campaigns. Theoretically, we distinguished between the organization/strategy aspect and the communicative focus of the candidates’ campaigning. The main empirical finding is, as expected, that candidate campaigning in Norway is party centred on both dimensions: In regard to the organization/strategy aspect, the candidates campaign as part of the party campaign organization and comply with the centrally decided campaign strategy. Candidates in Norway far from develop an independent campaign organization. They rely on resources from the central party and follow the centrally decided campaign strategy. In regard to the communicative focus dimension, the aim for most candidates is to attract as much attention as possible to their party and not to their own candidacy. However, we also found interesting differences between these two dimensions of candidate campaigning.

In the theoretical discussion we argued that placement on the list would probably influence candidate campaign behaviour. The empirical analysis reveals different effects of list placement for the two dimensions of candidate campaigning. In regard to organization/strategy, the top candidates emphasized the importance of the central parties to a greater extent than lower placed candidates. This indicates that the top candidates depended on the party for resources and assistance, and the in-depth interviews confirmed this. In regard to the communicative focus, it is the other way round. The top candidates were more prone to focus on themselves than lower placed candidates. However, individualized communicative focus was conditioned by party affiliation. On the one hand, all the Labour candidates were unanimous in saying that the aim of their campaign communication was to create as much attention as possible for the party. The top candidates in the Conservative Party and the Christian Democratic Party were much more inclined to focus on their own candidacy. These results are also in line with our expectations. Research on multilevel parties shows that there is less acceptance for a top-down culture in the Christian Democratic and the Conservative Party than in the Labour Party (Allern and Saglie, 2012). We relate this to the parties’ ideological and cultural background. The Labour Party was born and emerged outside parliament, while the Conservative Party was formed within parliament. Hence, the parliamentary face of the Conservative Party has always been more central to the party than that of the Labour Party, highlighting the role of the individual politicians. Earlier research has revealed that top-down culture is also emphasized in the Progress Party leadership (Allern and Saglie, 2012). However, the candidates from the Progress Party were less inclined to focus on the party than the Labour candidates. This indicates that the top-down culture is less integrated in the Progress Party’s organization than in Labour’s.

This difference between the communicative aspect and the organizational aspect of candidate campaigning is an essential finding, and shows the value of the conceptual distinction between these two aspects. Our results indicate that candidates can campaign as part of a greater party campaign organization and rely on the central party strategy, but still make themselves the focal point of their communication. This has consequences for our understanding of the nature of candidate campaigning in general: Higher levels of individualized campaigning in systems that for example opt for preference voting might be limited to the communicative dimension, and not the organizational and strategy dimension.

The conceptual distinction between individualization and localization also proved fruitful, as our results show that it is necessary to distinguish between localized and individualized campaigning, both conceptually and empirically. This should be a useful distinction even in plurality systems, where localized campaigns arguably more naturally centre on the individual candidate. Although the candidates with an individualized communicative focus were more likely to campaign on local issues, the general pattern is that most campaign on local issues. However, localized campaigning is also a matter of degree. Although candidates emphasize the importance of localizing the campaign, in the Norwegian context localizing is mostly about ‘translating’ the national campaign strategy to the regional or local level. Local parties and candidates strive to make the campaign message relevant for the voters in their constituency. A localized campaign strategy in Norway is not about independent local strategies or independent local party campaign organizations.

To conclude, some features of the Norwegian campaign environment – the nomination process, the electoral system and the strength of parties – curb tendencies towards individualization of candidate campaigning which is found in other systems and especially the US. Individualization is curbed in both organizational and communicative terms. Nevertheless, we have shown that there is room for variation with regard to individualized campaigning, and suggest that party ideological and cultural differences based on organizational traditions contribute to explaining these differences. In addition to the system traits discussed in this article, future comparative work must take these factors into consideration. To identify and refine such explanations should be a central aspect of future research on candidate campaigning.

Footnotes

Acknowledgements

We thank Hanne Marthe Narud, Elin Haugsgjerd Allern, Knut Heidar, Jo Saglie and two anonymous reviewers for their constructive and helpful comments and suggestions.

Funding

This research received funding from the Research Council of Norway.

Notes

References

Aalberg

Strömbäck

(2011) Media-driven men and media critical women? An empirical study of gender and MPs’ relationship with the media in Norway and Sweden. International Political Science Review 32(2): 167–187.

Aardal

(2011) The Norwegian electoral system and its political consequences. World Political Science Review 7(1): article 10.

Adam

Maier

(2010) Personalization of politics: A critical review and agenda for future research. In: Salmon

(ed.) Communication Yearbook 34. London: Routledge, 213–257.

Allern

Saglie

(2012) Inside the black box: Parties as multi-level organizations in a unitary state. West European Politics 35: 947–970.

Agranof

(ed.) (1972) The New Style in Election Campaigns. Boston, MA: Holbrook Press.

Balmas

Rahat

Sheafer

Shenhav

(forthcoming) Two routes to personalized politics: Centralized and decentralized personalization. Party Politics.

Bowler

Farrell

(1992) Conclusion: The contemporary election campaign. In: Bowler

Farrell

(eds) Electoral Strategies and Political Marketing. London: Macmillan, 223–235.

Brox

Shaw

(2006) Political parties, American campaigns, and effects on outcome. In: Katz

Crotty

(eds) Handbook of Party Politics. London: Sage.

Carey

Shugart

(1995) Incentives to cultivate a personal vote. A rank ordering of electoral formulas. Electoral Studies 14: 417–439.

10.

Dalton

Wattenberg

(2000) Parties without Partisans. Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press.

11.

Denver

Hands

Fisher

MacAllister

(2003) Constituency campaigning in Britain 1992–2001. Centralization and Modernization. Party Politics 9: 541–559.

12.

Duverger

(1954/1972) Political Parties. London: Methuen.

13.

Farrell

(1996) Campaign strategies and tactics. In: LeDuc

Niemi

Norris

(eds) Comparing Democracies. Elections and Voting in Global Perspective. London: Sage.

14.

Farrell

Webb

(2000) Political parties as campaign organizations. In: Dalton

Wattenberg

(eds) Parties without Partisans. Oxford: Oxford University Press, 102–128.

15.

Fisher

Denver

(2008) From foot-slogging to call centres and direct mail: A framework for analysing development of district-level campaigning. European Journal of Political Research 47: 794–826.

16.

Gallagher

Marsh

(eds) (1988) Candidate Selection in Comparative Perspective. The Secret Garden of Politics. London: Sage.

17.

Grofman

(2005) Comparisons among electoral systems: Distinguishing between localism and candidate-centered politics. Electoral Studies 24: 735–740.

18.

Heidar

Saglie

(2003) Predestined parties? Organizational change in Norwegian political parties. Party Politics 9: 213–239.

19.

Herrnson

(2007) Congressional Elections. Campaigning at Home and in Washington. Washington DC: CQ Press.

20.

Karlsen

Narud

(forthcoming) Nominations, campaigning and representation. How the secret garden of politics determines the style of campaigning and roles of representation. In: Esaiasson

Narud

(eds) In-between Election Democracy. Colchester: ECPR Press.

21.

Karvonen

(2010) The Personalization of Politics. A Study of Parliamentary Democracies. Colchester: ECPR Press.

22.

Kriesi

(2012) Personalization of national election campaigns. Party Politics 18: 825–844.

23.

Katz

Mair

(1995) Changing models of party organization and party democracy: The emergence of the cartel party. Party Politics 1: 5–28.

24.

McAllister

(2007) The personalization of politics. In: Dalton

Klingemann

(eds) The Oxford Handbook of Political Behaviour. Oxford: Oxford University Press, 517–588.

25.

Norris

(2000) A Virtuous Circle. Political Communication in Postindustrial Societies. Cambridge University Press, Cambridge.

26.

Plasser

(2002) Global Political Campaigning. London: Praeger.

27.

Poguntke

Webb

(2005) The Presidentialization of Democracy: A Study in Comparative Politics. Oxford: Oxford University Press.

28.

Swanson

Mancini

(1996) Politics Media and Modern Democracy. London: Praeger.

29.

Valen H Narud

Skare

(2002) Norway: Party dominance and decentralized decision making. In: Narud

Pedersen

Valen

(eds) Party Sovereignty and Citizen Control. Selecting Candidates for Parliamentary Elections in Denmark, Finland, Iceland and Norway. Odense: University Press of Southern Denmark, 169–215.

30.

Wattenberg

(1991) The Rise of Candidate-Centered Politics. Cambridge, MA: Harvard University Press.

31.

Whiteley

Seyd

(2003) How to win a landslide by really trying: The effects of local campaigning on voting in the 1997 British general election. Electoral Studies 22: 201–324.

32.

Zittel

Gschwend

(2008) Individualized constituency campaign in mixed-member electoral systems: Candidates in the 2005 German elections. West European Politics 31: 978–1003.