Marc Hooghe Sara Vissers

2008

Websites as a Campaign Tool for Belgian Political Parties. A Comparison between the 2000 and 2006 Local Election Campaigns.

pp. 171-196 in R. Davis, D. Owen, D. Taras, & S. Ward (Eds.), Making a Difference: A Comparative View of the Role of the Internet in Election Politics. Lanham: Lexington Press.
Lexington Press, 2008

The Belgian context

The Belgian party system can be considered as one of the most fragmented in the Western world. Partly this extreme form of fragmentation is due to the presence of a very open and proportional electoral system. Only in the year 2003 a new electoral threshold of five percent at the provincial level was introduced but thus far this threshold has not led to the actual disappearance of any of the small political parties (Hooghe, Maddens & Noppe 2006). A second element to explain this extreme fragmentation is the linguistic segregation of the party system. From the 1970s on, the former national parties split up in fully autonomous Dutch and French political parties. Although there is still some informal coordination going on between, e.g., the Dutch-speaking SP.A (Socialistische Partij) and the French-speaking PS (Parti Socialiste), in practice these parties function completely on their own. Given the fact that French-speaking parties do not compete in Dutch-speaking electoral districts, and vice versa, all parties run separate electoral campaigns. Only in the bilingual district of Brussels (ca. 10 per cent of the total population), both Dutch- and French-speaking parties compete with one another.

The overall result is that, at this moment, 13 political parties are represented in a federal or regional parliament in Belgium. For a population of a little more than 10 million inhabitants, this amounts to a very fragmented party system. This build-up of the party system has a number of consequences for the way political parties conduct electoral 1

campaigns. First of all, it has to be remembered that the two language groups remain completely segregated and do not compete with one another (Deschouwer 2004). Some authors even argue that Belgium does not have a single party system, but rather two completely different party systems: one for the Flemish region (ca. 6 million inhabitants) and one for the French-speaking region of Wallonia (ca. 4 million inhabitants). Second, given the small scale of the country and the large number of political parties, most politicians in Belgium remain convinced of the fact that personal contacts with the electorate are still possible, and indeed have the strongest impact. The electoral system, with relatively small districts, also offers incentives to maintain these personal and local relations with ones constituency. For all these reasons, especially an older generation of politicians initially was quite reluctant to adopt an anonymous medium like the internet for electoral campaigns. The campaign for the local elections of the year 2000 offered the first opportunity for Belgian political parties to use the internet in an election campaign. A study revealed, however, that internet use remained very limited. A number of parties and candidates launched a website as they had the feeling that they could not remain completely absent from this new medium. Maintenance of the websites, however, was extremely limited, and most websites did not offer much more information than an .html version of the leaflets and the brochures of the candidates and the parties (Hooghe & Stouthuysen 2001). The campaign for the local elections of 2006 offers us an excellent opportunity to ascertain the evolution since 2000. We have every reason to assume that parties will increasingly rely on the internet for campaigning. First of all, this is in line with an international trend (Lusoli, Ward & Gibson 2006). More specifically for Belgium, the percentage of the population with broadband access to internet has increased substantially during the past six years. Second, political parties in Belgium have become increasingly professionalized during this period. Following some corruption scandals in the 1990s, Belgian parliament has introduced a very generous system of financial allowances for political parties, allowing them to hire more professionals. In this article, we compare the internet presence of Belgian political parties in 2000 and 2006, with special attention to three research questions. First we try to ascertain to what

extent political parties use new ICT in a different way compared to the traditional communication channels? Second, we investigate whether all political parties use internet to the same extent? Is the gap between small and major political parties growing further or can we find some normalization? Finally we want to compare the situation in 2006 with the data coming from an earlier study in the year 2000.

Internet access in Belgium Our expectation that parties will have intensified their use of the internet as a campaign tool is based on the fact that the use of internet has risen substantially in Belgian society during the 2000-2006 period. Not only the number of connections has grown enormously but what is maybe more important is that the telephone line-based forms of access have almost completely disappeared in favor of broadband connections. This implies that consumers no longer pay per time unit, enticing them to spend longer hours on the Web, and it also means they can process more information and data. Partly because of aggressive marketing techniques by a number of large internet providers, broadband access is affordable to most households (current prices ca. 40 euro/month for a full service package) and according to Eurostat, Belgium and Estonia now have most household broadband connections of the entire European Union (Ottens 2006). An analysis of 2004 data also shows that Belgians are among the most active internet users in the European Union (Lusoli 2005). This trend is confirmed if we look at data from the Belgian association of Internet Providers (ISPA). While in the year 2000 there were 736,000 private internet connections in Belgium (one for every 13 inhabitants), in 2005 this was up to 1,771,000 (one for every six inhabitants). While in 2000 quite some households still relied on free telephone line-based access (with costs for every minute on-line), this form of accessing the internet has virtually disappeared in the year 2005 (Table 1).

Table 1. Internet Connections in Belgium, 2000-2005

2000 2005 Evolution Total private connection 736.483 1.771.016 + 140,5 % Free connection 394.766 209.649 -46,9 % Paid connection 341.717 1.561.367 + 356,9 % No broadband connection 287.741 47.842 - 83,4 % Broadband connection 53.976 1.513.525 +2704,1 % Total connections companies 132.533 416.007 +213,9 % Source: Internet Service Providers Association (ISPA 2005). Figures for July 2000 and 4th quarter 2005. Note: The data cover 97% of all the Belgian internet-providers

As the number of private internet connections does not inform us about the proportion of the population using the internet we also included survey data on individual internet use. Therefore we employed the European Social Survey data (2002, 2004), where we investigated the Belgian data. In the ESS the exact question is How often do you use the internet, the World Wide Web or e-mail whether at home or at work for your personal use? Comparing the data from the 2002 and the 2004 ESS confirms the trend that the use of internet is spreading very rapidly in Belgian society (Table 2). In 2002 34.7% of the Belgian respondents could be considered as frequent internet users (at least once a month), whereas in 2004 this percentage increased to 51.7%. As we already indicated, the Dutch and the French political parties can be seen as completely segregated party systems, and therefore it makes sense to make a distinction between the internet in the Dutch and the French communities of Belgium. While we see a small difference, for all practical purposes we can observe that both communities have adopted internet to the same extent. In this regard, both party systems are confronted with the same situation. Table 2. Regular private internet use in Belgium (2002-2004)
2002 n 2004 n Dutch 44.3% 1,200 52.5% 992 French 41.8% 643 50.2% 711 Total Belgium 43.7% 1,896 51.7% 1,776 Source: European Social Survey 2002, 2004. The internet use is measured as frequency of internet use for personal purposes. This variable is categorized between frequent users (at least once a month) and infrequent users (less than once a month). Distinction based on the language used most often at home of the respondent.

We can conclude, therefore that the internet market in Belgium is expanding very rapidly in Belgium, both in the Dutch as in the French part of the country. Because of specific

market conditions, broadband access is very widespread and theoretically this could function as an incentive for political parties to offer high tech content on their websites.

Research methodology Within the international literature, various methods have been proposed to analyze political website content in a systematic manner. In this chapter, we will mainly use the method develop by Gibson and Ward (2000). In this method, the focus is on two central aspects of political party websites, namely their purpose and function on the one hand; and their effectiveness in delivering these functions on the other hand. Initially this method was conceived of as a very elaborate coding instrument, involving fifty different criteria. The analysis of the party websites functioning was based on five primary goals of party websites: information, resource generation, campaigning, networking and participation. As various authors pointed out, the use of quantitative indicators in some cases posed serious problems concerning comparisons with other parties (Gibson and Ward 2000; Hooghe & Stouthuysen 2001). In a later version, Gibson and Ward (2005) simplified their method, by relying on a simple a dichotomous coding scheme (present=0 vs. not present=1). Moreover, they adapted the model to new developments in the ICTsector. In this chapter we use the 2005 version of the method, but we also added some new features, like the possibility of RSS feeds and the presence of Podcasting.

In coding the website content we followed the Gibson and Ward scheme quite closely, resulting in five main categories.

First, the website can be used for the simple provision of information, which amounts to a top-down process of distributing information from the party to individual users. This form of institutional communication is measured by coding the presence of information on the organizational history, party structure, party program, party statutes, media releases, speeches, whos who, etc. We also coded the presence of information on the party leader on the website, share in the profiles of the candidates, accessibility of the

information regarding the previous elections, an event calendar, frequently asked questions and an article archive or a library.

Second, parties can use their website to generate resources. This can apply to seeking donations, practicing merchandising and mobilizing users for party membership. Following Gibson and Ward we used an ordinal index for measuring this function, ranging from a simple listing of the postal address of the party (1), over the possibility to download a file to be posted (2) and other mixed download/postal combination (3) to a fully online transaction (4). Self-evidently, the higher a party scores on this ordinal index, the more innovative its use of new ICT means of communication.

Third, political parties can use their website as a networking tool. Networking can extend both to internal links (local candidates, local chapters, ideologically related parties), as to external links (to other parties or to non-political actors).

Four, websites can be used to allow or even to invite participation by citizens. Here we coded openness (presence of: e-mail address of the party leader, other politicians, party offices in general, named staff), feedback opportunity (presence of: (general) standardized massage forms, possibility to give feedback directly to the webmaster), interactive debate or comment opportunities (presence of: discussion forum, campaign/diary blog without comments, blog with an opportunity to give a comment, online questions and answers) and lastly the interaction with the site (presence of: opinion polls, games, petitions). Five, we coded how parties used their website for specific campaigning purposes, including targeting specific groups, download campaign materials etc. Separate from these main dimensions we also included a number of technical aspects of the websites (multimedia use, navigation tools, etc.). We limited our investigation to all political parties that are present in one of the parliamentary assemblees in Belgium (national or regional parliaments). This led to the introduction of eight Flemish parties, and five French-speaking parties (Table 3).

Table 3. Belgian Political Parties Name LanWebsite guage NL SP.A s-p-a.be NL Spirit spirit.be NL CD&V cdenv.be NL N-VA n-va.be NL VLD vld.be NL/FR vivant.be Vivant Vlaams Belang NL vlaamsbelang.org/be NL Groen! groen.be FR PS Ps.be FR MR Mr.be FR CdH lecdh.be FR Ecolo ecolo.be FR Front National frontnational.be

Ideology Socialist Progressive-nationalist Christian-Democrat Conservative Nationalists Liberals Liberal Extreme Right Greens Socialist Liberal Christian Democrat Greens Extreme right

% vote 2003

14.9 13.3 3.1 15.4 1.3 11.6 2.5 13.0 11.4 5.5 3.1 2.0

Parties included in this chapter. Name, language (Dutch= NL; French = FR), URL and ideology. Percentage of the vote in the 2003 federal elections. SP.A and Spirit formed a joint cartel for these elections. Election results: Ministry of the Interior.

In this research project we analyzed different layers of websites belonging to these URLs, using MetaProducts Offline Explorer to download the content of the websites. To supplement the content analysis, we interviewed the webmasters of different political parties, namely: Veerle De Gryze (SP.A), Pieter Verstraelen (Vivant), Steven Cryelman (Vlaams Belang), Piet De Zaeger (N-VA), Lieven Vanhollemeersch (Groen!), Patricia Van Overstraeten (CD&V), Pieter Montsaert (VLD), Angelo Callant (Spirit), Raphal Thimard (Ecolo), Nancy Vilbago (PS), Jean Paul Bastard (CDH), and Joanna Delaunoy (MR). The interviews were based on a semi-structured questionnaire and were carried out by telephone. The questions concerned practical issues of the activities of the webmaster and were focused on the following aspects: the budget spent on the website, the number of persons responsible for updating the site (full-time vs. part-time), the use of a content management system, cooperation with ICT companies, forms of cooperation at the national and local level, the average number of unique visitors per day, the time and localization specification of the website visitors, directories to a website (direct via URL of the website, via search engine, etc.) and the characteristics of the decision making process on the content and design of the website. In comparison with the research

conducted in 2000 (Hooghe & Stouthuysen 2001), the webmasters were much more willing and/or able to indicate the number of visitors in a more reliable manner. On the other hand, we found it difficult to obtain information on the cost of website management, as the parties often did not know themselves the exact amount spent on it, or did not include this as a separate line in their budgets. In order to address the first research question on how parties use ICT, website content was analyzed and it was ascertained to what extent political parties actually use the new communication opportunities being created by the internet. We can distinguish three forms of communication via internet (Hooghe & Stouthuysen 2001). First, the pureinteractive communication is based on a dialogue between non-institutionalized actors, without any interference from institutionalized actors. A political chatroom would be a typical example of such interaction. The second form is institutional communication. Here, an institutional actor (e.g. a political party) provides top down information toward other actors (individual or institutionalized), without any possibility for feedback or interaction. Offering information on a website by a political party or the government can be seen as a clear example of this communication form. The third form can be seen as a mixture of the two former types and is called an institutional-interactive communication. Again, the initiative comes from the institutionalized actor, but at the same time non-institutionalized visitors have the possibility to interact with the institutionalized actor. The opportunity to comment on website content would be a typical example of this communication form. In this third setting there is no possibility for a simultaneous interaction between two non-institutional users of the website. Mostly, they do not even know from each other how they reacted. In this case we can speak of a sequential interaction (Gibson & Ward 2005). Our basic assumption for this analysis is that the more the parties use the interactive opportunities being created by new ICT, the more democratic potential these new technologies have.

Content analysis of political parties websites

Information provision Almost all political parties use their websites to provide information about their party to potential voters, explaining their point of view on various policy issues (Foot & Xenos 2005). Only two parties, Vivant and Front National, two minor parliamentary parties obtain a remarkable low score on the information provision function. Vivant presents only the party program, the event calendar, the media releases and the related archive on its website. On the website of the Front National one only finds media releases. The party program has been removed from the Front National site following the decision of the Brussels Court of Appeal (April 2006) that the party leader is guilty of inciting racism, which is an illegal act according to Belgian law. All parties present an overview of the most recent press releases on their websites. Moreover, they give access to article archives or to a library, enabling users to find older press releases or articles. With the exception of Vivant and FN, all parties have an event calendar and a kind of whos who about their party leaders and representatives. It is striking here that, in line with the party ideology, the Groen! website offers very little information on individual Members of Parliament. It only lists their names and contact options, but no additional information on activities or other individual actions is being provided. Larger parties more often provide information about the history of the party, the structure and statutes, and a detailed profile of their representatives. Frequently asked questions and electoral information about the previous elections appear only on a few websites. Almost all parties try to avoid to give too much attention to the person of the party leader and often one even has to look to find relevant information on the party president. There are only two exceptions to this pattern: both for the PS as for the MR, there is a highly visible special index link to the website of the party president. This is no coincidence: while most often a party president cannot be a minister in a government, the PS president is the prime minister of the Walloon regional government, and the MR president is the vice prime minister of the Belgian federal government. Apparently, this very strong power position also translates itself into a specific position on the party website.

Comparing the results from the current analysis, conducted in 2006, with the results from the analysis in the year 2000 (Hooghe & Stouthuysen 2001), shows that little has changed with regard to the provision of information. We do see, however, that parties are slowly finding more imaginative ways to provide information on their websites. While in 2000 it was still customary that the entire text of the speeches of the party president was being reproduced on the party website, this practice seems to have been abandoned by the year 2006. The results from both observation years show that parties see their websites as an important medium for the quick dissemination of a large amount of information. In general, it is safe to argue that information provision by far remains the most important function of party websites in Belgium. It seems clear that parties remain much more interested in website as a tool for institutional communication than in using the interactive possibilities created by the new ICT technology.

Resource generation If we take a look at the resource generation function, it turns out that all the parties use the opportunity to reach out to (potential) members online, but again to varying degrees. All the parties, with the exception of three minor parties: Vivant, Ecolo and Front National, provide an enquiry form on their websites. While all parties made some attempts to attract members online, only three of them offer the possibility to buy merchandise and only two parties explicitly solicit donations. Only the two liberal parties (VLD and MR) actively use their website to sell merchandise online. Again, maybe it is not a coincidence that these are liberal parties, that in general are quite supportive of private enterprise. They present pictures of their goods on the websites and enable online transactions. Vlaams Belang is the only other party selling merchandise online, but without providing an online transaction (payment by bank transfer). The Flemish Christian-Democratic party CD&V provides a list of books, but for more information it refers to other websites, e.g. of the publisher, local division, personal website of a politician, etc. The Flemish Green party Groen! gives the possibility to subscribe to their monthly review by providing the bank account number.

10

Only two parties explicitly solicit donations, Groen! and Front National. Both parties are quite small, and because of their scale they are largely excluded from the Belgian system of party financing. Therefore they are more dependent on donations than any of the larger parties. Groen! and the Front National list their bank account number and an email enquiry form on their websites. Comparing with our observations from the year 2000, does not reveal any remarkable differences. In comparison with other countries, Belgian political parties remain quite modest in soliciting donations or other forms of financial support. Largely this is the result of the official system of party financing, so that political parties indeed are not dependent on financial support from the public. The effort to generate resources usually remains limited to offering an online membership form.

Table 4. Information Provision and Resource generation by Belgian Political Parties

SP.A Information provision Organizational history Party structure Party program Party statutes Media releases Speeches Whos who Leader focus Candidate profiles Electoral information Event calendar Frequently asked questions Article archive or library Resource generation Donation index Merchandise index Membership index CD &V VLD VB Vivant Spirit GRN NVA PS MR CdH ECOL FN

X X X X X X X X

X X X X X X X X

X X X X X X X X X X X X X X X X X 10 X

X X X X X X X

X X X X X X X X X

X X X X X X X X X

X X X X X X X

X X X X X X X

X X

X X X

X X X X

(X) X

X X X

X X X 11

X 9

X 10

X 10

X 4

X 9

X 6

X 10

X 9

X 9

X 6

X 2

4 4

1 4 5

4 4 8

3 4 7

3 3

4 4

3 1 4 8

3 4 4 4 4 4 4 8 4 4 3 3 3 6

11

Networking To determine the networking function of the party websites, we coded internal and external links. The coding makes clear that all parties devote a lot of attention to networking. All parties with the exception of the small Vivant and Front National placed links to the personal websites of their politicians, local parties and other internal groups on their websites. Table 6 lists the number of links to local party and candidate websites, for every party. The figures show that the practice of linking to other party-related websites is much more widespread among the Dutch language parties than among the French language parties. Also, larger parties (obviously with more elected officials) provide much more links than the smaller parties. Moreover the number of links with local chapters and with candidate websites also informs us about the nature of the party structure. A striking example is the SP.A, with lots of websites by local chapters and relatively few personal websites. This conveys the message that the SP.A is a rather strictly organized party, where the local chapters are encouraged to adopt a nationally developed format. On the other hand, local politicians receive little leeway in the party to develop their own personal campaigns and ideological profiles. Christian-Democrats and Liberals on the other hand are more individualized political parties, with stronger political personalities.

12

Table 5. Networking and Interaction by Belgian Political Parties

SP.A Networking Internal candidates Local parties Other internal groups External partisan links Reference links Commercial links Opponents Participation Openness email leader Email other politicians Email party office Email named officials/Staff Feedback General feedback Standardized message form Interactive debate/comment Discussion for a Diary/blog (without comments) Blog with comments Online q&a Interaction with site Opinion polls Games Petition CD &V VLD VB Vivant Spirit GRN NVA PS MR CdH ECOL FN

X X X

X X X

X X X

X X X

X X X

X X X

X X X

X X X

X X X

X X

X X X

X X

X X X

X X

X X

X X

X X X 6

X X X 6

X X X 6

X X X 5

X X X 6

X X

X X

X X

X X

X X

X X

X X

X X

X X

X X

/ /

X X

X X

X X

X X X

X X X

X X

X X X X X X X 5 7 5 4 5 4 4 X

X X X X 6 X 5

If we consider external links, all parties except Vlaams Belang provide links to ideologically related organizations, like sister parties in other EU countries, affiliated organizations, etc. Furthermore, all parties except CD&V en Vlaams Belang provide links to politically neutral websites such as news broadcasts, newspapers, government sites, national libraries etc. CD&V and Vlaams Belang are the only parties which provide a link to a commercial publisher.

13

We can observe a distinct patterns with regard to providing a link to other, competing parties. In Flanders, this is clearly not a common practice and we fink these links only on the N-VA website. In Wallonia, on the other hand, all parties do so with the exception of the small Front National. The interviews with the party webmasters, however, revealed that this is largely a symbolical linkage. Very few visitors actually enter the websites by entering from a link provided by a different political party. There seem to be very little people interested in surfing from one party website to another.

Table 6. Links to local party chapter and candidate websites

SP.A CD&V VLD Vlaams Belang Spirit 51 14 N-VA 105 14 Groen! 108 1 Vivant 1 PS 46 64 MR 91 67 CdH 16 Ecolo (83) 5 FN 3 Note: Belgian has 589 local councils: 308 in Flanders, 262 in Wallonia and 19 in Brussels. Theoretically, Flemish parties could have 327 local chapters; French parties 281. Local party website 312 158 158 102 Personal Website 48 133 97 24

Interaction If we take a look at the overall score on the participation index (Table 5), we notice that all political parties perform rather weak. Obviously parties do not pay much attention to providing discussion forums, blogs or online dialogues. Again, as in the year 2000, allowing for interactive communication clearly is not a priority for political parties. Only six parties (three from each language group) provide some platform allowing for interactive debate and comment. In order to get a better insight into the character of the virtual discussions, we examined one (recent) discussion on each of these six websites more in depth (Van Selm, Jankowski & Tsaliki 2002). What is striking is that discussions are usually well managed (Table 7). The discussion is initiated by a statement from a

14

Member of Parliament or another politician, and most parties succeed in managing the discussion quite efficiently, ensuring that most contributors remain on topic.

Table 7. Political party web-based discussions

Subject SP.A Monopolistic behavior of brewery multinational Vivant Dependency on oil supplies N-VA Should railways become a regional competence? 28 28 . 27 0 0 Citizens discussing a statement posted by a politician 14 -comments sometimes off-topic -emotional Informal MR Ethics and politics Ecolo Democracy in the Democratic Republic Congo 14 14 10 3 1 Blog (politician) with comments (citizens) 8 -comments restricted to topic blog

Number of contributions posted by: -MP -citizens . -men -women -gender not known Formula

18 18 . 16 0 2 Citizens discussing a statement posted by a politician

106 6 100 101 0 5 Citizens discussing a statement posted by a citizen

19 19 17 2 0 Guest (politician) answers questions by users 15 -a broad variety of topics/ questions informal

Participants -MP -citizens Quality

Tone speech

of

11 -comments restricted to statement -well composed contributions Informal

1 17 -comments restricted to statement -well composed contributions informal

informal

The PS gives the possibility to post some humoristic comment. This platform is not a real discussion forum, but it gives the opportunity to state a private opinion, criticize or praise something, etc. The MR organizes a chat session (via webcam) with a special guest (politician) on a regular basis. Citizens are encouraged to send their questions also in advance. This communication form is more an online Q&A than a real debate between citizens and a member of parliament. In Flanders, too, on-line discussion is not the major function of the party websites. Two smaller parties (Vivant and N-VA) are most active in this regard by offering discussion forums on different issues. Moreover, both parties provide a platform for online Q&A. Pieter Verstraelen, the webmaster of Vivant,

15

admitted that at first party officials feared that the forum would be used by opponents for all-out attack on the party (cfr. Foot & Xenos 2005), but in practice this seldom occurred. In Flanders, SP.A is the only major party that enables online debates and comments via a large variety of blogs on different issues. All party members are allowed to start a new blog by posting a new statement. The figures in Table 7 suggest that women are practically absent from these online discussions. In the Flemish online discussions all contributions were posted by men, and on the French sites we counted only a few contributions by users with a female name. By itself this is a troubling finding: if party elites in some way or another get the impression that the on line discussions are in any way representative for the opinion of their voters or their constituency members, this in practice means that only the male members or voters actually are being heard, while women do not receive any voice at all. CD&V provides a weekly diary without giving the possibility to the visitors to react. The CD&V webmaster told us that the party used to have a discussion forum during the last three years but they decided to remove this from the website. The reason was that too many people used the forum for undesired goals and it was too difficult and timeconsuming to control what happened on the forum.

It is striking that in 2000 parties apparently invested more in maintaining chatforums or other interactive discussions (Hooghe & Stouthuysen 2001). Already then, however, several webmasters voiced complaints that these discussions easily led to rude behavior and damaging actions by opponents of the party. Apparently, parties changed their tactics during the pas six years. We see less emphasis on a public dialogue now than we did in the year 2000, and users have fewer possibilities now to post comments that can be read by every other user. Parties do keep up the effort to create some kind of interactive communication, however, but less so in public. Users receive the e-mail address of the party leader and other politicians, as well as the personal (named) e-mail addresses of the party officials. This asynchronous communication form was already popular in 2000 and is remains the most popular from of communication on the 2006 websites. From the perspective of the party, it has the advantage that it allows for interaction and communication while at the same time the party itself controls the flow of

16

communication, avoiding any possible embarrassing web content. If political parties exploit the interactive possibilities of internet, they opt for what Hooghe and Stouthuysen (2001) have called an institutional-interactive form of communication, where they still control the flow of the information. In addition to the e-mail addresses, most of the parties, except Vlaams Belang, PS and Front National, offered also a standardized message form. Half of the parties gave the possibility to send a message directly to the webmaster. E-mail addresses and message forms are not the only way to encourage online participation. Some of the websites use features that did not fit into the coding scheme, e.g. they encourage visitors to react by telling them that their opinion is important for the party (send us your idea, time for action, contact us). The aim of it is to convey to users the idea that the party cares about the opinions of ordinary citizens.

Campaigning All parties undertake some forms of online campaigning, though in general Belgian parties score rather low on these issues in comparison with other countries (Bimber & Davis 2003). This is in line with the general observation that election campaigns in Belgium are conducted in a rather low-key, non-aggressive manner. Negative advertising, e.g., is completely absent in Belgian electoral campaigns, for the simple reason that the fragmentation of the party system inevitably leads to coalition governments, also on the local level. Parties, therefore, are discouraged from creating hostile feelings toward their competitors, but also potential coalition partners. All parties except Front National provide the possibility to join an e-mail update list and to download various text. Five Flemish parties and two Walloon parties give the opportunity to download promotional material, such as television clips, posters, cartoons, desktop wallpaper and the party logos. Some parties provide the html code for placing advertisement and the party logo on other websites. On the website of Groen!, Vlaams Belang, PS and MR, one can send e-cards to other internet users. During the analysis, we could not find any indication for targeting expatriate voters and/or marginal constituency. Apparently, parties do not take full advantage of the new possibilities which multimedia create for an online campaigning. In practice, old-

17

fashioned forms of campaigning (bulletin boards, leaflets, personal visits, public meetings, ) still play an important role in Belgian electoral campaigns, partly because of the small scale of the country. Table 8. Campaigning and Multimedia Use by Belgian Political Parties
SP.A Campaigning Negative campaigning Targeting expatriates Targeting marginal constituency Join an email list Become online volunteer Send a link/epostcard Download literature Download screensavers/banners Download promotion materials Multimedia factor Sound Video Live streaming Flash SMS RSS Access Text only option Text only documents to download/print Foreign language translation Blind/visually impaired software Navigation Navigation tips Search engines Fixed menu bar on lower level pages Site map/index Freshness Update daily CD &V VLD VB Vivant Spirit GRN NVA PS MR CdH ECOL FN

X X

X X

X X X X X 4 2 X X X X

X X X

X X

X X X X X

X 4

X 4 2 2 3

X 3

X 4 4 2

X 4

X X

X 3

X 3

X 1

X 3

X X

X X

X X

X X

X X

X X

X X

X 1 2 2 2 2 2 2 2 2 1

X 2 1 X X

X X X 3 1

X X X 3

1 1

1 1 1 1

1 1

X 2 1

1 1

X 2

X 2

X 3 1

18

Multimedia While nearly all parties present pictures on their websites (photo galleries), only six out of thirteen provide video clips. In general, the available video clips cover events and speeches. We can see a clear division between the Flemish and Walloon parties. Four out of five French speaking parties have video clips on their websites, whereas on the Flemish side only SP.A and Vlaams Belang do the same. Other websites limit their multimedia use to moving icons. Not a single party provided live streaming and the use of sms. In our coding scheme we added the new upcoming technique: Really Simple Syndication. Four parties already give users the possibility to use this new method. The RSS feeds provide web content or summaries of web content together with links to the full versions of the content as an XML file. The feeds facilitate indication and allow a website's frequent readers to track updates on the site by using an aggregator.

Accessibility All parties clearly do an effort to make their website accessible, even for users relying on telephone access, by providing text only documents. The option to have the entire website as a text-only version, however, that was still available in the year 2000, has completely disappeared. The Flemish and French Christian-Democrats are the only parties who offered special software for the blind and visually impaired users, which is in line with their ideology on social inclusion. The results are mixed with regard to helping the users to navigate the website. Small websites like Vivant or Front National hardly need any navigation tips. But even for the major parties, there is not always a website map available and also the presence of search engines seems to be a random pattern.

19

Summary of content coding Table 9. Summary Codings for Website Content

SP.A Information provision Resource generation Networking Participating Campaigning Multimedia/glitz factor Access Navigation Freshness Total score 9 4 5 5 4 3 1 3 1 35 CD &V 10 5 4 7 4 VLD 10 8 5 5 2 VB 10 7 4 4 4 3 2 1 35 Vivant 4 3 3 5 2 Spirit 9 4 5 4 2 1 2 1 1 29 GRN 6 8 5 4 3 1 2 2 1 32 NVA 11 4 6 8 3 PS 10 4 6 6 4 MR 9 8 6 5 4 2 1 2 27 CdH 9 4 5 5 2 2 2 3 32 ECOL 6 3 6 3 4 3 1 3 29 1 13 FN 2 6 3

2 1 1 34

2 1 1 34

2 1 20

2 1 1 36

2 2 34

When combining all the previous codings into a summary table (Table 9), it is clear we do not find all that much variation between political parties. Although the coding is to some degree arbitrary, it should be noted that all the major parties receive a total score of something between 32 and 35. It has to be remembered that all of these are major parties, receiving a substantial amount of government. Apparently this allows them to maintain an online website presence. It is just as clear that these major parties share the same priorities: they devote a lot of attention to providing information, while they score lower on offering various interaction possibilities. Pure forms of interactive use are sometimes completely absent.

The smaller political parties, however, have but a few members of parliament and as a consequence they also receive less state funding. Their online presence clearly is more limited. Only the Front National experiences difficulties in maintaining a visible web presence, but this might also be due to the dubious legal status of this extreme right party.

Information from Interviews

We used the interviews with the webmasters to obtain more information on the actual development and maintenance of the websites. Almost all parties left the development and technical management of the website to an outsourced company. Cooperation

20

between the party headquarters and the outsourced company is based on a content management system. Within this system the party is only responsible for the content and the update of the website. The Flemish green party had to terminate the contract with their outsourced company due to financial problems: in 2003 this party lost most of its state funding following a severe electoral defeat. Only three parties (CD&V, Vivant and Vlaams Belang) developed their website themselves and they are also responsible for maintenance and updating. We also asked how many people were professionally involved with the websites, but the answers on this question were a bit dubious. It does seem clear, however, that within the larger and well-funded parties at least one full time staff members is responsible for the website (web master). The smaller parties, in general do not have a full time author. Most of the time this is solved by making the authors of the several texts and messages responsible for posting content on the website. Smaller parties like Vivant and Groen! also use volunteers to take care of website content and updating. It seems clear that money is a dividing line here. Major parties can manage their website presence in a very professional manner, but this is not necessarily the case for the smaller parties.

Internet campaigning is a special challenge for local elections, since local candidates and party chapters do not always have the resources available to use all the new ICT technology (Strachen 2003). If we take a look at the cooperation between the national and local level within the party, some differences in autonomy and responsibility can be noticed. Major parties like SP.A, CD&V and VLD use a system of content management for the local sections and these sections are encouraged to use the same format. One of the webmasters, Patricia Van Overstraeten of CD&V, explained that the national party headquarters offers a package with layout, banners and national information. The local chapters do not have to do anything than to add this with their own local information. According to Ms. Van Overstraeten, this system is used because most local party chapters do not have sufficient resources to develop a website on their own.

As we already mentioned, the interviews did not result in a clear estimate of the costs involved in maintaining an internet presence. Nevertheless, we do see clear differences

21

between the main political parties and the smaller ones. SP.A indicates that they spend 4,200 euros a month on their website. 4,000 euros are paid for maintenance of the website, while there are also 200 euros hosting costs. CD&V, on the other hand, uses its own serve and therefore does not receive a monthly invoice from an ICT company. The part however, employs a full-time webmaster and here too the total personnel cost could be estimated at something like 5,000 euro every month. The smaller parties have less money to spend on website presence. Spirit estimates that the party spend no more than 4,000 euros a year on the maintenance of the website. Vivant is a special case because the founder and president of the party actually owns a major ICT company, and there is not a clear financial division between those two activities. Groen! indicates that after the severe financial cuts the party suffered following the 2003 electoral defeat, the party has to limits itself to a simple update of the website, without any money for additions or restructuring the websites. All webmasters indicate that the party websites are updated daily, especially with regard to the press releases. The major parties estimate that website content is being refreshed up to six times a day.

In 2000, there was still not a common standard available for counting the success of a website. At that time, some webmasters mostly counted the number of hits, sometimes even the numbers of pages visited, and others counted the number of unique visitors. In 2006, however, this problem seems to have been solved as most webmasters could provide us with information on unique visitors. As could be expected, the number of visitors has grown exponentially between 2000 and 2006. In 2000 Vivant attracted on average 30 visitors a day, in 2006 this number was up to 350. For CD&V the same number increased from 200 to 1600 visitors per day. VLD measured 300 visitors per day in 2000, but the webmaster could (would?) not inform us about the number of visitors in 2006. SP-A has in 2006 (March) on average 2822 visitors per day. As the figures in Table 10 make clear, the extreme-right Vlaams Belang claims to have by far the most-visited website of all political parties in Belgium. We cannot control these figures, that were supplied by the party itself and which might strike it us as out of balance compared to the electoral appeal of this party. On the other hand, the Vlaams Belang is by far the most outspoken and the most contested party in Belgium and it tends to polarize public

22

opinion. So theoretically it might be possible that more people visit its website, compared to the website of the more mainstream political parties. Table 10. Unique visitors per day (May 2006)
SP-A 2.822 CD&V 1.600 VLD VL BL 7.456 Vivant 350 Spirit 800 Groen! 473 N-VA 2.272 PS MR CDH Ecolo 557 FN (1000)* Source: Webmasters of these parties, interviews April/May 2006. Unique visitors are counted only once no matter how many times they visit the site (per day). Unique visitors are identified with their IP number. * webmaster couldnt tell whether or not unique visitors.

Conclusion

In this article we wanted to document the way Belgian political parties use the internet as part of their communication and campaigning strategy. It seems clear that Belgian political parties rely on a mix of various media. Given the fact that 2006 is a year of local elections, and given the small scale of the country, parties obviously still believe in small scale contacts with their constituency, using traditional media. The internet is an addition to this campaign mix, especially for political parties catering for a relatively young audience, but clearly it does not replace any of the older campaign tools. If we consider the functions of the websites, it turns out that little has changed since the elections in 2000. Political parties use their websites primarily for a fast distribution of information, in a one-way and top-down communication process. Opportunities for interaction and dialogue are relatively scarce, and most often they are conducted in such a way that the party controls the flow of information. If we would establish a ranking, it is clear that for political parties the institutional flow of information remains more important than any of the interactive forms of communication that are offered by the internet. In general, the extent of the campaigning function of the websites remained 23

limited to the possibility to update an e-mail list, to download information and promotion materials. This is line with the general observation that election campaigns in Belgium might be professionalized, but in general they are still quite traditional compared to the US or the UK (Gibson & Rmmele 2001). Multimedia use is very limited, and this is related to the fact that political parties are not allowed to buy commercial television spots. While in other countries this video clips can also by distributed on internet, this is not the case in Belgium. Resource generation is very limited, given the fact that parties can rely on a generous system of state funding.

While parties tend to offer the same kind of information on their websites, we also see some clear differences, revealing distinct party features. Just from the structure of the party website, e.g., it is already clear that the presidents of the PS and MR enjoy a very strong power position, which is not the case for other party presidents. The SP.A reveals itself to be a well-organized but rather centralized structure, which is not the case for VLD or CD&V. The Christian-Democrats on the other hand, practice their inclusive ideology by offering information to visually impaired internet users.

One of the major concerns about the use of internet campaigns is that major political parties will be able to monopolize campaigning, because they have more resources available to develop and maintain attractive websites. Smaller parties, on the other hand, risk to be marginalized since they do not have the mains to use all the new communication techniques that have become possible in the internet era. The Belgian results, however, do not really confirm this negative hypothesis. Only for two of the smallest parties, Vivant and Front National, it is obvious from the website that they have few resources and that they find it difficult to maintain an attractive website. Furthermore, since activities from Members of Parliament and of ministers usually make up most of the website content, for these parties it apparently is sometimes even difficult to find enough current information to fill the website. For all the other parties, however, it seems as if they can actually manage a real website presence. We can observe that the updates are much more regular on the website of the large parties than on those of the small parties, but there is certainly not a linear relation

24

between size of the party and attractiveness of the website. Here much seems to depend on the personality and the enthusiasm of the web masters and their technical staff. It is also clear that the parties reach out to different audiences. The Christian-Democrats, e.g., traditionally attract older voters, and their use of the website is kind of mainstream and modest. Spirit, Groen! or SP.A, on the other hand, attract a younger electorate, and apparently they are more willing to take a risk with regard to the layout of their websites, which would appeal a younger electorate. Our Belgian findings, therefore, suggest a threshold model with regard to the relation between party size and website presence: once a party has crossed a certain threshold with regard to resources, it can maintain a viable website presence. The answers from the webmasters suggest that a 40,000 euro/year budget seems an absolute minimum amount required to make daily updates on a websites. Parties falling below the threshold have a hard time to produce a viable and up-to-date website. For parties above that threshold this task is easier, and we certainly do not have an indication that the more you spend on a website, the more attractive it becomes. In Belgium, 11 of the 13 parties studied clearly have sufficient money to spend, because they are included in the system of state funding. Systems of party funding might not solve all the problems, but they obviously can make it easier for even relatively small political parties to join the era of online-campaigning.

25

References

Bimber, Bruce & Richard David (2003). Campaigning Online. The Internet in U.S. Elections. Oxford: Oxford University Press. Deschouwer, Kris (2004). Political Parties and Their Reactions to the Erosion of Voter Loyalty in Belgium, pp. 179-206 in Peter Mair, Wolfgang Muller & Fritz Plasser (eds.). Political Parties and Electoral Change. London: Sage. Foot, Kirsten & Michael Xenos (2005). Politics as Usual, or Politics Unusual? Position Taking and Dialogue on Campaign Websites in the 2002 U.S. Elections. Journal of Communication, 55(1), 169-185. Gibson, Rachel & Andrea Rmmele (2001). Changing Campaign Communications. A Party-Centered Theory of Professionalized Campaigning. Harvard International Journal of Press/Politics, 6(4), 31-43. Gibson, Rachel & Ward, Stephen (2000). A Proposed Methodology for Studying the Function and Effectiveness of Party and Candidate websites. Social Science Computer Review, 18(3), 301-319. Gibson, Rachel, Ward, Gibson & Lusoli, Wainer (2005). UK Parties and the 2005 Virtual Election Campaign: Not Quite Normal? paper presented at the Parties and Election Campaigning Online Workshop, Oxford Internet Institute, University of Oxford, 20-21st October, 2005. Hooghe, Marc, Bart Maddens & Jo Noppe (2006). Why Parties Adapt. Electoral Reform, Party Finance and Party Strategy in Belgium. Electoral Studies, 25(2), in press. Hooghe, Marc & Patrick Stouthuysen (2001). Het politiek gebruik van internet naar aanleiding van de gemeenteraadsverkiezingen van 8 oktober 2000. Res Publica, 43(4), 507-528. ISPA (2005). Statistics on Internet Access in Belgium. Brussels: Internet Service Providers Association. Lusoli, Wainer (2005). A Second-Order Medium? The Internet as a Source of Electoral Information in 25 European Countries. Information Polity, 10, 247-265. Lusoli, Wainer, Stephen Ward & Rachel Gibson (2006). (Re)connecting Politics? Parliament, the Public and the Internet. Parliamentary Affairs, 59(1), 24-42. Ottens, Morag (2006). Use of the Internet among individuals and enterprises. Brussels: Eurostat. Strachan, J. Cherie (2003). High-Tech Grass Roots. The Professionalization of Local Elections. Lanham: Rowman and Littlefield. Van Selm, M., Jankowski, N. & Tsaliki, L. (2002). Political parties onine: DigitalDemocracy as reflected in three Dutch political party web sites. Communications, 27, 189-209.

26