ON PEERS AND AGENT CHOICE

In this dissertation I study the impact of decision by peers (others) on decision making of a focal agent. I study this effect in two broad marketing contexts a) influencer marketing b) charitable giving. In my first essay, I study how regulatory punishment on a set of social media influencers affects endorsement strategy of other influencers in the same regulatory environment. In my second essay, I study the effect of donation behavior of other donors on the focal donor. Specifically, I investigate the role of popularity of a charitable cause on donations of new donors and existing donors.
ESSAY 1 – Regulatory notices and endorsement disclosure Social media platforms such as Instagram have become an essential channel for influencer marketing. Regulatory bodies such as FTC (in the US) and ASA (in the UK) require influencers on these platforms to declare an advertised social media post as an ad using hashtags such as #ad, #sponsored. However, often influencers fail to disclose the endorsements. To discourage these unprofessional practices, FTC sent warning notices to 90 influencers in March 2017. We use this event as a quasi-natural experiment setup to estimate the impact of FTC notices on a) influencers’ disclosure levels and b) follower engagement. We curated a novel dataset that consists of nearly 150 thousand Instagram posts over a 6 years period. We find that advertising disclosures increased for the influencers who received the notice, and their follower engagement (likes and comments) was adversely affected. Furthermore, we estimated the deterrence effect of FTC notices on other influencers. We find significant spillover effects on other influencers in the FTC jurisdiction. Specifically, the disclosure percent of the influencers who did not receive notice also increased compared to the control group. Our findings provide valuable insights to regulators and social media managers on the direct and deterrence effects of regulator notices.
ESSAY 2 – Popular or crowded: Subscription-Based Donations Subscription-based donations are becoming a popular fundraising tool as they are perceived to yield a high donor lifetime value. A common practice of online donation platforms is to display, for each cause (e.g., cancer treatment or education provision), the donor group size (number of people donating to that cause). We use data from a subscription-based donation platform to study the effect of displaying donor group size on new donors and current donors. We use a) repeat donations of individual donors and b) an exogenous shock to the platform that shifts the donor group size to identify its impact on the two donor groups. We find that displaying the number of donors can act as a double-edged sword — encouraging new donors (a "bandwagon" effect) while discouraging existing donors (a "bystander" effect) from subscribing. We suggest the managers be careful about displaying the number of donors as the net effect on subscriptions can vary with the "life cycle" of the charity and its donors. Specifically, managers can leverage this information when new donors signup but should not disclose this information to current and active donors.