The Wisdom of Crowd-Funders: What Motivates Cross-Border Private Development Aid?

Report | December 2013

 By: Raj M. Desai and Homi Kharas

BROOKINGS

(Introduction and conclusions, for full report follow the link at the end of the text)

INTRODUCTION

In 2010, foundations, nongovernmental organizations (NGOs), religious groups and other charitable organizations in the United States contributed $39 billion to international development causes (Hudson Institute 2012). By comparison, $30 billion in US official development assistance (ODA) was disbursed during the same year. For US-based organizations, this represented a doubling of international private, voluntary development assistance over the past decade.

In recent years, the proliferation of private development aid has been facilitated by peer-to-peer (or “crowd-funding”) practices. As with the broader “crowd-sourcing” phenomenon which solicits information from large numbers of individuals for various entrepreneurial activities, crowd-funding platforms bundle large numbers of small, individual contributions for investment, grants or loans. The bundling of funds is generally done through internet-based social networks. From the United States, internet-based companies such as Global Giving, Kiva, Wokai and Zidisha have channeled millions of dollars to individuals and partner organizations in developing countries.

Despite the tremendous growth in private development assistance of all kinds — from mega-charities to “micro-philanthropy” — very little is known about the allocation and selectivity of private aid. Compared to official aid, private aid — whoever provides it — is obviously more sensitive to the preferences of philanthropic-minded individuals who determine allocations across countries and, within countries, across sectors, projects and individuals. More importantly, crowd-funding philanthropy affords an opportunity to test a central premise behind arguments for expanded private aid — namely, that private aid avoids the political and strategic considerations that influence bilateral ODA allocation, and better matches recipient need with individual donor preferences.

Information on the allocation of crowd-funded private aid, and on the choices made by private citizens who contribute to international causes, can potentially reveal the implicit preferences of philanthropic citizens in a way that cannot be captured by looking at official aid allocations. There are several possibilities: that crowd-funders behave in accordance with rational-choice theories of charity, allocating money based on individual-specific preferences; that crowd-funders behave like official donors, responding to a combination of recipient-country need, expected performance and commercial and/or strategic value; or that crowd-funders make allocative decisions on the basis of group behavior and norms.

This article presents new data on crowd-funded development assistance, which allows us to analyze the motivations behind individual contributors. We analyze data from Kiva, the largest provider of crowd-funded microcredit to developing nations. We develop a model to show how the allocation of aid through crowd-funding websites reflects the preferences of philanthropically-minded citizens regarding development assistance, and then use data on Kiva’s transactions to examine empirically the factors that affect the supply of private development aid, as well as to determine the extent to which private preferences differ from official aid agency allocative mechanisms.

We argue that the rate at which individual microloan requests are funded by Kiva’s community of lenders, once they are posted on the Kiva website, can be interpreted as a proxy for crowd-funder preferences regarding private development assistance. We can therefore use survival analysis of the time to fund each project to estimate the significance of a number of covariates.

We find that Kiva’s crowd-funders are generally not influenced by the usual set of official aid determinants (including foreign-policy considerations, recipient-country poverty and recipient-country institutional quality). Additionally, Kiva crowd-funders do not appear to base lending decisions on the usual indicators of credit risk. We find, instead, that the type of diaspora and migrant networks of aid recipients in the crowd-funder’s country is a stronger determinant of hazard rates, and that the nature of associational networks and social linkages between prospective private donors and aid recipients will more likely affect crowd-funded aid allocation than recipient-country or project risk.

CONCLUSIONS

We examine indirect evidence of the funding preferences of private aid givers using data on crowd-funded microloans through Kiva, an internet-based peer-to-peer platform that bundles contributions into loans for entrepreneurs in developing countries. The behavior of private aid givers has been little examined to date, despite the increase in private development aid. We identified three possibilities: that crowd-funders act as risk-averse charitable contributors, making funding decisions based on project-specific risks and incentives; that crowd-funders behave like official foreign aid agencies, funding projects based on a combination of country need and institutional quality; or that crowd-funders are motivated by the social networks connecting them to countries in which projects take place. Because Kiva crowd-funders expect to have their principal repaid, Kiva’s project data affords us an opportunity to test the risk-aversion of crowd-funders; because Kiva’s projects span some 80 developing countries, we can also examine whether aid flows through Kiva are as selective as official development assistance.

Using survival analysis of funding rates for Kiva projects between April 2006 and December 2010, we find weak support for the view that crowd-funders are risk-averse with respect to microcredit in developing countries. Kiva’s crowd-funders do prefer loans to women, as well as those of shorter duration and smaller amounts. However, they reject the group liability approach of traditional microfinance and only weakly prefer lending through lower-risk partner microfinance institutions. We find almost no consistent support for the perspective that crowd-funders act selectively toward projects based on the poverty or institutional quality of the country.

By contrast, we find strong support for the argument that crowd-funding is essentially an expressive act that enables individuals to “connect” with micro-entrepreneurs, much in the same way that individuals can “sponsor” children in developing countries through a number of NGOs. In this regard, we find that the nature of social relations that developing countries are able to rely upon in richer countries — in particular, through their communities of migrants — has a strong effect on the funding rates of Kiva projects. Kiva crowd-funders prefer to lend to countries which claim larger numbers of more recent, wealthier immigrants, and from which large number of refugees also come. For two reasons, we speculate that this is not necessarily due to participation by immigrants themselves as crowd-funders. First, Kiva lending moves in the opposite direction of remittances, which would be unlikely if immigrants themselves were largely responsible for Kiva funding to their home countries. Second, countries with large percentages of immigrants employed in the high-wage financial services sectors receive money at slower rates, the opposite of what would be expected were immigrants themselves participating as crowd-funders. Rather, we suggest that immigrant communities, through their social ties with native-born populations, provide information about their home countries to prospective crowd-funders.

These findings have implications for official aid policy.12 In contrast to years past, the collective-action costs of private aid appear to be minimal, especially with the proliferation of internet-based crowd-funding platforms. Moreover, internet technology appears to have reduced the advantage that official agencies once held in terms of minimizing the transaction costs of disbursing aid. Finally, private aid now has significant advantages over official aid in avoiding agency costs, as private aid givers can give money to recipients in developing countries in a much more direct way. Indeed, the rapid growth of crowd-funded private aid may be attributed to the attractiveness of this “short route” to giving.

Not all recipient countries, however, are organized to take advantage of this spread of private aid. Another obvious conclusion is that aid recipient countries would do well to organize themselves to take advantage of new forms of private aid. For example, in India, MFIs must first obtain approval from the Reserve Bank of India before they can borrow abroad — an obvious barrier to accessing private loans from Kiva. Our findings also suggest that the design of projects can be fine-tuned to make them more attractive to donors. To give an example: It is probably more effective to invest in providing assistance to entrepreneurs to allow them to develop project ideas than to invest in building the capacity of microfinance intermediaries. Private lenders seem not to care too much about the rating of these agencies.

The phenomenal growth of internet-based giving is testimony to the potential for private aid to reach a scale which can be significant in global terms. What has not been shown is that organizing aid in this fashion is more effective for development. A comparison of development effectiveness between public and private aid platforms is an important direction for future research.

For the full report follow the link: http://www.brookings.edu/~/media/research/files/reports/2013/12/crowd-funders-developoment-aid-desai-kharas/wisdom-of-crowdfunders-v4.pdf