Page 153 - 臺大管理論叢第32卷第2期

NTU Management Review Vol. 32 No. 2 Aug. 2022

B = 1.015, p < 0.01). In sum, our empirical results remain robust to this alternative model
specification.

4. Discussion

In this study, we examine funding dynamics on the crowdfunding platform by pin-
pointing the role of the bystander effect. While previous studies have provided some evi-

dence on this relationship (Chan et al., 2020; Mollick, 2014), our study contributes to this
research stream by explicitly examining the notion of the “bystander effect” on the crowd-
funding platform. By using granular day-by-day campaign data from flyingV, we investi-
gate whether the bystanders would influence the daily campaign fundraising performance.
Drawing on insights from social psychological research, we find that a higher bystander

ratio discourages these supporters from funding the project. Consequently, a higher by-
stander ratio may lead to a lower level of daily pledge amount. Our study suggests that
investors’ funding decisions are influenced not only by the project rewards themselves but
also by the presence of bystanders.

Furthermore, we examine the boundary conditions of the bystander effect in the
crowdfunding context (Fischer et al., 2011; Hussain et al., 2019). First, we find that project
legitimacy, derived from firm incumbency, is an important signal to bystanders; our
results reveal that project legitimacy mitigates the negative bystander effect. Moreover,

the longer the duration of the fundraising period, the weaker the negative bystander effect.
Specifically, when the project duration is extended, crowd funders have more opportunities
to collect campaign-related information, thus reducing the information asymmetries and in
turn converting bystanders into funders.

One unique feature of the flyingV platform is that it records the number of supporters
and backers. Based on this information, we derive the number of bystanders by subtracting
the number of backers from the number of supporters. Such information is important since
most crowdfunding platforms such as Kickstarter only record the number of backers who

have contributed funds to the project (Kuppuswamy and Bayus, 2018). Using data from
the flyingV platform, we can capture information about the number of bystanders who are
attentive to the project but have not contributed funds yet. Therefore, data from flyingV
provides a unique opportunity to capture the bystander effect in the crowdfunding context

145