Online Political Campaigning During Malaysia’s GE15: Blurred Lines
Published
The advent of account-based influencers during Malaysia’s recent general election does not portend well for the country’s political landscape.
Malaysia’s 15th general election (GE15) witnessed an emerging trend which saw candidates hiring account-based influencers to run political advertisements on their platforms. They tweeted and retweeted the candidates’ tweets about their political messaging and their campaign activities such as walkabouts and speeches. Through an analysis of the Twitter feeds of these account-based influencer accounts, as well as semi-structured interviews with the account owners, it was found that these influencers charged candidates in the same way they charged small and medium-sized businesses to run advertisements.
This is significant, as it blurs the lines between astroturfing and political campaigning. Astroturfing refers to the practice of artificially inflating the volume of chatter on the ground, particularly surrounding an opinion or a topic being discussed. The practice lacks transparency, as it is not made known to people that the tweets are advertisements and do not evince genuine support for specific political candidates.
Account-based influencers have Twitter accounts with large followings and they run advertisements on these accounts. Unlike traditional influencers who base their brand on a specific person, account-based influencers brand themselves based on their account. Many account-based influencers organically accumulate large followings through various means such as posting relatable quotes or humorous content. After attaining a huge follower base, they monetise their ability to pull an audience by running advertisements for businesses. However, account-based influencers are often anonymised or based on a person who gives away little personal information about themselves. The account owners are not obliged to reveal their real names, faces, or identities outside their platform. Traditional influencers based on specific persons subject themselves to intense public scrutiny. One example is @surayaror, an influencer with 65,000 followers, who monetised her platform in various ways, including advertising for robo-investor companies. But account-based influencers have less accountability to their followers. One example is @pelabur_bijak, an account-based influencer with little to no details about the account holder. The account has 40,400 followers and shares finance-related knowledge and advertises affiliated Shopee links.
Astroturfing refers to the practice of artificially inflating the volume of chatter on the ground, particularly surrounding an opinion or a topic being discussed. The practice lacks transparency, as it is not made known to people that the tweets are advertisements and do not evince genuine support for specific political candidates.
Political advertising by account-based influencers blurs the line between astroturfing and political campaigning by political candidates. During GE15, political candidates hired several account-based influencers to tweet and retweet tweets that supported them and inflated the volume of chatter about their campaigns. Many voters would come across these tweets without knowing if the tweets that reached their timelines reflected genuine support for the candidates or were paid advertisements.
One reason why Malaysian political candidates may be drawn to such advertising practices is that it is highly affordable and accessible, as seen in Figure 1. If a politician chose to engage with Account A, they would potentially have access to a 37,700-strong audience for as low as RM 5 (US$1.35) per tweet. Thus, a budget of merely a few hundred ringgit would allow for hundreds of tweets to flood the timelines of thousands of voters on Twitter and would inflate the volume of perceived support for the politician.

One case study saw a Barisan Nasional candidate paying account-based influencers to tweet and retweet tweets using his tagline as a hashtag for his campaign (the candidate contested in Johor and lost). As seen in Figure 2, the hashtag was used 352 times. However, it was only used by 33 different individuals. This shows that the same few Twitter users were repeatedly using the hashtag — at a frequency of an average of 10.7 tweets per user over the campaigning period. The tweets had a reach of 343,000 different followers that belonged to these 33 individuals. Upon closer analysis, many of those who tweeted the hashtag more than twice were account-based influencers, and almost all were promoting their advertising services and platform on their Twitter accounts. Two of the account-based influencers in Figure 1 were among them. It is not made clear whether all these influencers truly supported the candidate or were paid to promote his campaign.
Through conversations with some of these account-based influencers, they shared that their line of work is in a grey area because it is similar to what cybertroopers do, even if they themselves are not cybertroopers. Cybertroopers focus exclusively on spreading political propaganda, while account-based influencers propagate political advertisements along with other types of advertisements. Furthermore, these influencers’ accounts are not fake accounts that were created solely to propagate political messages, unlike cybertrooper accounts. In fact, some of the influencers made conscious choices in deciding whether to take on such work. For example, one influencer declined an offer to work for a certain party because their ideology did not align with his beliefs.

Despite the efforts put into running political advertisements and engaging such influencers on Twitter, they did not guarantee electoral success in GE15. In this study, all three politicians who had used account-based influencers lost in the constituencies they were campaigning in. There are many reasons for this. For example, previous research has shown that social media has the ability to improve the impression of a politician but many more factors are in play when it comes to swinging the actual votes: such as support for the policies the politician is championing, and whether an even stronger candidate is contesting. Nonetheless, the Philippines and Indonesia are good case studies in exemplifying what may happen to the online information and political landscape should astroturfing and political influencer-led advertising be allowed to run rampant. In both countries, influencers made lucrative amounts of money for their exclusively political services, which motivated the influencers to spread disinformation in order to receive high engagement numbers. This will sow the seeds for a toxic political environment which, over time, will erode trust in the electoral process, leading to greater voter cynicism and disengagement with politicians.
2023/28
Amirul Adli Rosli was a Research Officer at the Regional Social and Cultural Studies Programme, ISEAS – Yusof Ishak Institute.