The Questions of Giving: Crowdfunding for the Poor

Project: Research

View graph of relations


Description is the largest crowdfunding platform that allows funders to make interest-free loans topeople in poverty worldwide for their small business projects. Kiva crowdfunding is actuallyintermediated by Kiva’s field partners, which act as microfinance institutions (MFIs). They vetand post projects on Kiva, manage raised money, monitor projects and repay funders throughKiva. Field partners charge high interests rate (an average of 34.65%) while they get interest-freeloans from Kiva funders. In 2012, Kiva launched a new channel, Kiva Zip for direct interest-freeP2P funding. On Kiva Zip, people can support borrowers from the US or Kenya. There is no fieldpartner, and a borrower only needs an endorsement from a Kiva-recognized third party, whichnevertheless is not obligated to any responsibility for the loan. The borrower pays no interest rate.The introduction of Zip changes the market structure of the crowdfunding platform and may affectthe behavior of participating parties. The proposed study intends to leverage the naturalexperiment setup to empirically examine the impact of the direct channel on the behavior changeof funders as a start. We first identify which type of funders are more likely to lend on Zip. Thenwe investigate how funders change their contribution on Kiva to demonstrate the cross-channelimpact of Zip on Kiva. We also briefly discuss our plan to investigate how the new channel affectthe intermediaries’ operations and our analytical approach to capture the strategic interactionamong funders, borrowers, field partners, and the platform and to explicate the social welfareimplications of introducing a new channel, although not in detail.We draw on the theories of altruism to describe funder incentives to contribute. We argue thatthey are motivated with a mix of pure altruism – sole care for others’ welfare, and the utility theyderive for themselves from the altruistic action per se, the “warm glow”. We explicate how the twochannel structure may affect funders’ incentives and what types of funders are more likely tocontribute to the direct channel. We also identify the cross-channel changes of funders’contribution caused by the introduction of the new direct channel. We propose that if the fundersare highly motivated with pure altruism, the direct channel may cannibalize funders’ contributionfrom the intermediated channel. For funders with more concern on sustaining their contributionfor “warm glow” utility, the direct channel stimulates their contribution on the intermediatechannel, where project risk reduces as the direct channel segment low quality projects away.We collect data from Kiva and Zip. We derive funders’ loan portfolios and devise two indicatorsto measure funders’ pure altruism level, average interest rate and poverty level of the portfolio.We hope to establish that a funder with more pure altruism is more likely to contribute on Zip.Then we leverage the natural experiment design to identify the causal effect of Zip introductionon funders’ contribution change on Kiva to test for the cannibalization or stimulation impact. Tofortify the findings, we will also use a difference-in-difference specification. Our study sheds lightson how prosocial funders select contribution channels and the cross-channel spillover. We mayalso provide evidence about funders’ altruistic incentives being a mix of pure altruism and utilityderived from the altruistic action. The significance of “warm glow” incentive suggests that theplatform cannot only rely on people’s pure care for others or social recognition, but also needs tocontrol project risk to sustain funders’ contribution.Although not described in detail, we also propose to examine the changes in intermediaryoperations and performance after Zip introduction. We have already collected data of MFIs’operations and performance and can identify operational changes of partner MFIs after Zip.Moreover, we propose to develop an analytical model to further explicate the strategic play of thethree parties in the two channel platform structure. We hope to clarify in equilibrium how themarket is segmented with different types of funders and borrowers and the social welfare andpolicy implications of having two different channels. We hope to clarify the conditions underwhich the direct channel is viable. The results can provide important insights on howcrowdfunding platforms can design their channels and maximize the social welfare of the system.


Project number9042576
Grant typeGRF
Effective start/end date1/09/1724/06/20