Social Media, Money, and Politics: Campaign Finance in the 2016 US Congressional Cycle

Abstract With social media penetration deepening among both citizens and political figures, there is a pressing need to understand whether and how political use of major platforms is electorally influential. Particularly, the literature focused on campaign usage is thin and often describe the engagement strategies of politicians or attempt to quantify the impact of social media engagement on political learning, participation, or voting. Few have considered implications for campaign fundraising despite its recognized importance in American politics. This paper is the first to quantify a financial payoff for social media campaigning. Drawing on candidate-level data from Facebook and Twitter, Google Trends, Wikipedia page views, and Federal Election Commission (FEC) donation records, we analyze the relationship between the topic and volume of social media content and campaign funds received by all 108 candidates in the 2016 US Senate general elections. By applying an unsupervised learning approach to identify themes in candidate content across the platforms, we find that more frequent posting overall and of issue-related content are associated with higher donation income when controlling for incumbency, state population, and information-seeking about a candidate, though campaigning-related content has a stronger effect than the latter when the number rather than value of donations is considered.

Keywords: Social Media, Federal Election Commission, Congress, Election, Donation, Campaign Finance.

Introduction Scholars have sought to understand the relationship between technology and democracy since the 1990s (Barber, 1998). With rapidly rising adoption of social media by citizens, US politicians are increasingly aware of the power of major platforms to communicate and organize for political purposes (Gainous and Wagner, 2014; Margetts, John, Hale, & Yasseri, 2015). The growth in social media campaigning specifically has been mirrored 2 by growth in literature analyzing usage itself and implications for a range of electorally- related outcomes (Boulianne, 2015). As social media penetration continues to deepen among the American electorate, there is pressing need to determine whether and how political candidates’ use of these platforms has electoral significance. Social media adoption is widespread for any protocol, with 70% of US adults on Facebook and 20% on Twitter as of 2016. Empirical work has demonstrated a rise in citizen and candidate interaction on the main platforms. Roughly 40% of Americans had posted and 80% had seen political content on social networking sites (SNSs) as of April 2014. Most importantly, followership of political figures on the main SNSs is on the rise; while 14% and 24% of 18-29 year olds and 6% and 21% of 30-49 year olds followed elected officials, political parties, or candidates for office in 2010 and 2014 respectively (Anderson, 2015), 35% of the American online population now does so (Kalogeropoulos, 2017). The practice of followership is bipartisan, with supporters of both parties equally likely to follow political figures on social media (Anderson, 2015). Studies seeking to explain increased campaign usage distill unique offerings of major social media platforms. Most, primarily focused on Twitter, argue such sites facilitate the “most inexpensive, unmediated, and closely focused forms of communication in campaign history,” (Gainous and Wagner, 2014, 54); further, these platforms are ideally suited to the types of messaging in which office-seekers want to engage, as they enable candidates to create succinct themes and highlight victories rather than explain the minutiae of complex legislation. It has been suggested that although the major social media platforms were not originally created for political purposes, the fact that they are low cost, allow direct communication with the public, and provide access to a wide body represent advantages over traditional phone, mail, and website-based campaigning (Auter and Fine, 2017). Because social media has been shown to be fundamentally different from ‘campaigning as usual’ (Bode et al, 2016), the implications of rising use of social media in campaigning are worth investigation. Cognizant that social media are increasingly commonplace, a plethora of studies have begun to describe usage and test impact across a range of electoral areas, such as vote gains (Yasseri & Bright, 2016; Bright et al., 2017), political participation (Bode, 2012), and political learning (Dimitrova et al., 2014). These studies offer a wealth of information regarding the ways in which online campaigning is playing a rol

…(Full text truncated)…