Essays on the Impact and Mechanism of Crowdfunding

關於眾籌機制及其影響的研究

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date21 Aug 2019

Abstract

Enabled by the Internet, crowdfunding has emerged as a new mechanism of fundraising by gathering small amounts from many funders. By connecting funders and fund-seekers worldwide, it helps parties that are overlooked by traditional financial organizations to access capital. This dissertation examines this new mechanism of crowdfunding and its impact, and it consists of three essays.

The first essay, “Crowdfunding for Microfinance Institutions: The New Hope?” focuses on the impact of crowdfunding on the borrowing organization, which is the traditional organizations in microfinance industry, and is, the microfinance institutions (MFIs). The MFIs, whose social mission is poverty alleviation, provide small and uncollateralized loans to poor populations and serve clients that are often excluded from commercial banks. Some crowdfunding platforms share the same social mission as MFIs and focus on prosocial projects. They provide a channel for MFIs to obtain funds on the basis of a project. We investigate the impacts of using crowdfunding on the performance of MFIs and find that MFIs’ operational self-sufficiency performance improves, and the interest rate decreases after joining crowdfunding, implying the improvement of MFIs’ performance. We further explore the underlying mechanisms of this improvement. The empirical evidence shows that the enhanced operational self-sufficiency and lowered interest rate are mainly the results of efficiency improvement.

In the second essay, “Group-based Reputation and Prosocial Lending,” we study the effect of group-based reputation on prosocial contributions in crowdfunding. This study focuses on the group-based reputation system, a motivation mechanism, on prosocial crowdfunding platform. Two theoretical lenses, social identity theory and public good theory, result in contravening predictions and are adopted to understand group-based reputation. Our empirical analysis suggests a crowd-out effect and that group-based reputation (top-10 team rankings) serves as a signal of other members' contributions, which leads to a decrease in subsequent team contribution. Our further analyses provide evidence for the signal and crow-out effect. The theoretical and practical implications of the study for crowdfunding, online prosocial contribution, and platforms are discussed.

The third essay, “Implicit Racial Discrimination in Crowdfunding,” focuses on the racial discrimination issue on crowdfunding platforms. Racial discrimination has long been recognized as an obstacle for minorities to access capital in traditional credit markets. While crowdfunding creates new fundraising opportunities for fund-seekers who are underserved in the traditional financial market, there could be potential issues of racial discrimination that affects its financial inclusion because funders can freely choose the projects they support. Our study investigates whether racial discrimination exists in prosocial crowdfunding and tries to understand whether it is derived from implicit forms of discrimination. Based on the suggestion that exposure to positive or negative instances of African Americans activates people’s implicit racial attitudes, we exploit prominent positive and negative news events concerning African American as manipulations to test the effects of implicit discrimination on an online crowdfunding platform. Our analysis shows that, on average, funders contribute less to African American loans. A prominent negative news event about African Americans increases racial discrimination in that, after the news event, African American loans become even less likely to be funded. Conversely, a prominent, positive news event about African Americans reduces discrimination, such that African American loans have an even higher chance to get funded.

This dissertation contributes to the literature and practice in several ways. First, by investigating how using crowdfunding affects the borrowing organizations, the first study identifies the causal effect of crowdfunding on MFIs and provides empirical evidence for the real impact of crowdfunding, and particularly prosocial crowdfunding. To our knowledge, this is one of few studies that focuses on the impact of crowdfunding. Specifically, we find that using crowdfunding as an alternative financing source helps MFIs improve sustainability (around 3%) and reduce interest rates (around 5%). Our further findings on the underlying mechanism show that the improvement of MFIs does not directly result from the decrease of capital cost but the improvement of operational efficiency and risk management, which is derived from the information disclosure and crowd-monitoring on crowdfunding platforms. Specifically, they decrease the operational cost (14%), personnel cost (10.4%), impairment expense (24.5%), and portfolio at risk (15.2%). In addition, we draw on the partnership between crowdfunding platforms and borrowing organizations (MFIs) and collected data from multiple sources. Data from the borrower side is rare, particularly the longitudinal data before and after using crowdfunding. Our findings can help crowdfunding platforms enhance their design to promote the success of borrowers and provide references for microfinance industry that crowdfunding could be a promising prescription.

Second, we reveal the effect of group-based reputation mechanisms on prosocial behavior in a prosocial crowdfunding context. Previous studies have focused on the role of individual-based reputation in prosocial contexts and shown the positive impact. Online platforms, such as crowdfunding, enable the users’ communities and the related group-based motivation mechanism. We explore and answer the important theoretical and practical questions about how the group-based reputation system affects prosocial behavior, whether the group-based reputation has a similar positive effect of individual-based reputation and motivates larger groups of users or if it works differently compared to individual reputation. Our findings provide empirical evidence that group-based reputation mechanisms do work in a different and more complicated way. Group-based reputation reflects the contribution of other members and leads to a crowd-out (negative) effect. Specifically, the subsequent amount of contribution is relatively lower (around 38%) on average when the lending team achieved the top-10 list. Our findings have significant implications for platforms that rely on prosocial contributions from (e.g., prosocial crowdfunding) users in terms of using group-based reputation mechanisms. These may need to be more carefully designed because the crowd-out effect may be triggered. Information transparency and designs that enhance the signal of others’ contributions and group identities may help make the reputation systems more efficient.

Third, our study is one of the first to empirically investigate implicit discrimination in crowdfunding, thus answering the call for more empirical studies of implicit discrimination (Neumark 2018). Traditionally, economists have focused on the explicit form of racial discrimination, that people consciously discriminate, including taste-based versus statistical-based discrimination. However, discrimination can also be automatic and outside of the discriminator’s awareness. Such an implicit form of discrimination can play an important role in driving people’s decisions, and the related intervention of debias is different. Our findings first present the evidence of racial discrimination in prosocial crowdfunding by showing that the average amount contributed by funders is relatively smaller for African American loans compared to White loans (around 6.5%) on average. Our further results provide evidence of implicit discrimination. Our findings also have practical significance by offering platforms low-cost solutions to reduce racial discrimination by embedding exposure to related, positive instances in the designation of platform.