An Empirical Analysis of Crowdfunding
Thomas Lambert * Université catholique de Louvain
Armin Schwienbacher * Université catholique de Louvain & University of Amsterdam Business School
This study investigates characteristics of crowdfunded projects and drivers of success. In line with the community view of crowdfunding, our results indicate that much of the funds provided are either donations or are entitled to receive a final product created by the project, rather than equity or cash payments. Moreover, crowdfunding initiatives that are structured as non-profit organizations tend to be significantly more successful than other organizational forms, even after controlling for various project characteristics. This finding is in line with theoretical arguments developed by the contract failure literature (e.g., Glaeser and Shleifer, 2001) that postulates that not-for-profit organizations may find it easier to attract money for initiatives that are of interest for the general community due to their reduced focus on profits.
This version: March 24, 2010
* Contact details of authors: Louvain School of Management, Université catholique de Louvain, Place des Doyens 1, 1348 Louvain-la-Neuve, Belgium; email@example.com & firstname.lastname@example.org. We are grateful to all participants in this survey. We wish to thank Paul Belleflamme and Ulrich Hege for their helpful comments and also Grégoire Krieg for his excellent research assistance. Any remaining errors are the authors alone.
Electronic copy available at: http://ssrn.com/abstract=1578175
1. Introduction It is well recognized that new ventures face difficulties in attracting external finance at their very initial stage, regardless whether bank loans or equity capital (see, e.g., Cosh et al., 2005). While business angels and venture capital funds fill gaps for larger amounts, the smallest amounts are provided by entrepreneurs themselves and the 3Fs (friends, family and fools). Still, many ventures remain unfunded, partially because of a lack of sufficient value that can be pledged to investors, partially because of unsuccessful attempts to find and convince investors. Recently, creative founders1 have made use of a new source of finance – so-called crowdfunding – by tapping the crowd instead of specialized investors. The concept of crowdfunding finds its root in the broader concept of crowdsourcing, which uses the “crowd” to obtain ideas, feedback and solutions in order to develop corporate activities. In the case of crowdfunding, the objective is to collect money for investment; this is generally done by using social networks, in particular through the Internet (Twitter, Facebook, LinkedIn and different other specialized blogs). The crowd-funders (those who provide the money) can at times also participate in strategic decisions or even have voting right. In other words, instead of raising the money from a very small group of sophisticated investors, the idea of crowdfunding is to obtain it from a large audience (the “crowd”), where each individual will provide a very small amount. For instance, artists seeking money for their new album through SellaBand sell participation rights at the price of $10 each. Once 5000 are sold, the artist can approach a record company that will then start producing the CD. While crowdfunding has been primarily used in the entertainment industry so far (especially music and movie), a few initiatives have been undertaken recently in other industries such as journalism (Spot.Us), beer (BeerBankroll), software (Blender Foundation) and fashion (Cameesa). Even more surprising, the amounts that have been raised through crowdfunding have continuously increased, reaching more than £ 1 million by Trampoline Systems for the financing of the commercialization stage of their new software. To take another example, BeerBankroll was able to raise $ 2.5 million from its 5000 members. These recent entrepreneurial experiences in raising capital through crowdfunding raise new and interesting questions. For instance, what affects the success chances of entrepreneurs to reach their capital targets through crowdfunding? What drives the crowd to participate in the first place? Does a
In this study, we instead use the term “entrepreneurs”. With this term, we encompass all the individuals and organizations using crowdfunding.
Electronic copy available at: http://ssrn.com/abstract=1578175
promise to reward the crowd-funders or give them voting right make fundraising more successful? And from a conceptual perspective, how does crowdfunding distinguish from crowdsourcing? In this paper, we first discuss a definition of crowdfunding and several issues pertaining to the practice of crowdfunding in connection with entrepreneurial activities. Crowdfunding leads to complexities that are distinct from its overarching concept, namely crowdsourcing. Next, we derive characteristics of crowdfunding initiatives by means of hand-collected data of 51 initiatives. These data are helpful in providing a better understanding of how such initiatives are structured and what motivates them. Perhaps surprisingly, only a limited fraction of initiatives is based on donations. The major fraction are passive investments; i.e., investments with a promise of compensation but no direct involvement in the decision-making process or provision of time or expertise for the initiative. In most of the cases, the compensation is to receive a product or service from the financed activity (e.g., BeerBankroll gives tangible benefits such as a t-shirt from BeerBankroll, gift cards to popular retailers, and memorabilia). Shares are offered in one third of our sample. Finally, we examine what drives their chances of success. This is done through multivariate analysis. A striking result is that non-profit associations are significantly more likely to achieve their target level of capital in comparison with other organizational forms (corporation, individual or in connection with a single project). This result appears robust to different econometric specifications. This finding is consistent with the notion that non-profit associations find it easier to attract capital from donors and other sources, since their focus is not purely profit-driven. As shown theoretically, among others, by Glaeser and Shleifer (2001) and Ghatak and Mueller (2009), profit-driven organizations may be prone to focus too much on profits at the expense of other dimensions such as quality of the product or service provided. This in turn may not be desired from donors and other sources aimed at fostering specific initiatives.2 The remainder of the paper is structured as follows. Section 2 offers a definition of crowdfunding and how it relates to concepts such as crowdsourcing and open-source practices. Section 3 presents the related literature. Section 4 discusses data collection and defines the variables used. Section 5 presents key characteristics of crowdfunding initiatives, based on our hand-collected dataset. Section
A related strand of literature argues that individuals may provide public goods due to social reputation, which induces pro-social behavior (Bénabou and Tirole, 2006). Moreover, experimental economics studies indicate that individuals become discouraged when faced with fines in case of underperformance or when treated unfair (Falk and Kosfeld, 2006), indicating that monetary incentives may at times deter individuals to undertake initiatives and behave altruistically.
6 provides results of a multivariate analysis on determinants of crowdfunding success. Finally, Section 7 concludes by suggesting topics for future research.
2. A Definition of Crowdfunding As mentioned, the concept of crowdfunding can be seen as part of the broader concept of crowdsourcing, which uses the “crowd” to obtain ideas, feedback and solutions in order to develop corporate activities.3 The term “crowdsourcing” has been first used by Jeff Howe and Mark Robinson in the June 2006 issue of Wired Magazine, an American magazine for high technology. Kleemann et al. (2008) point out that “crowdsourcing takes place when a profit oriented firm outsources specific tasks essential for the making or sale of its product to the general public (the crowd) in the form of an open call over the internet, with the intention of animating individuals to make a [voluntary] contribution to the firm's production process for free or for significantly less than that contribution is worth to the firm.” Taking Kleemann’s et al. (2008) definition as starting point, several caveats and clarifications need to be made in order to transpose the definition of crowdsourcing to crowdfunding. Hereafter, we offer a discussion on the application of this definition to crowdfunding; we ultimately provide key elements in understanding why crowdfunding is embedded in the definition of crowdsourcing. Raising funds by tapping a general public (or the crowd) is the most important element of crowdfunding. This means that consumers can volunteer to provide input to the development of the product, in this case in form of financial help.4 From this perspective, crowdfunding is a subset of crowdsourcing, since the latter encompasses also financial help. Several platforms have emerged recently, such as Fundable, Kickstarter, Kiva, Sandawe, and SellaBand. These intermediate between entrepreneurs and potential crowd-funders. Therefore, a distinction can be made between direct and indirect fundraising because at times entrepreneurs make use of such crowdfunding platforms instead of seeking direct contact with the crowd. These platforms at times share some similarities with online lending markets5 (Everett, 2008; Freedman and
For a non technical introduction of crowdsourcing, see Howe (2008). We note that an important motivation for relying on crowdsourcing is that it may contribute in reducing production costs (Kleemann et al., 2008). For instance, the pharmaceutical company Innocentive has organized its crowdsourcing practice in form of a tournament, where the provider of the best solution was rewarded with a prize (Albors et al., 2008). 5 These platforms are sometimes called online peer-to-peer lending.
Jin, 2010); while the latter more prominently target social entrepreneurship, crowdfunding platforms have a broader scope of entrepreneurial initiatives. As pointed out by Brabham (2008) and Kleemann et al. (2008), among others, the development of Web 2.0 is a critical ingredient that has facilitated the access to the “crowd”.6 Roughly speaking, Web 2.0 is a Web-as-participation-platform7 that facilitates interaction between users. This structure is crucial for entrepreneurs to be able to easily reach networks of investors or consumers. Through a case study, Larralde and Schwienbacher (2010) highlight the importance of efficient communication and networking. They argue that this is an inherent component of any crowdfunding process. The argument is also in line with the study of Lee et al. (2008), who identify three properties of Web 2.0 that enhances the ability of entrepreneurs: openness, collaboration, and participation. In contrast to the Internet that existed before the bursting of the dot-com bubble, the more recent Web 2.0 technology allows user to provide content (beyond simply reading existing one), interact with each other and thereby create value for the company (Lee et al., 2008). While the use of the Internet to make an “open call” may be very efficient for crowdsourcing in general, it can become more problematic for crowdfunding, especially if it involves the offering of equity to the crowd. Indeed, making a general solicitation for equity offering is limited to publicly listed equity. Companies cannot do a general solicitation, unless they received prior authorization from their national securities regulator. In many countries, there is also a limit as to how many private investors a company can have. For instance, Media No Mad could not have more than 100 shareholders, as imposed by French law (Larralde and Schwienbacher, 2010). While the crowdfunding process of this company was made in the public domain, shareholder contracts for the purchase of shares were however only signed with 100 individuals, as a way to overcome these legal problems. This creates important legal limitations to crowdfunding initiatives, given that the input of the crowd is capital and not an idea or time. In the case of Trampoline Systems, the company was required to prepare a detailed mechanism in order to avoid any problems with the UK financial markets regulator. Therefore, most initiatives do not offer shares but provide other types of rewards such as a product or membership. Crowd-funders make voluntary financial contributions with or without the expectation of receiving compensation. This can take various forms, including cash, bonds, stocks, profit sharing and preordering of products. At times, this can be accompanied by voting rights or other active involvement
In spite of the fact that ventures could be theoretically tapping the crowd via other means of communication than those associated with Web 2.0, we think that it could be marginal and does not exist to our knowledge. 7 Refer to O’Reilly (2007) for an in-depth understanding of Web 2.0.
in the crowdfunding initiative. Our empirical study will provide evidence on different types of rewards and rights, as well as the magnitude of the financial contributions generated through crowdfunding. In practice, entrepreneurs relying on crowdfunding may combine it with other forms of crowdsourcing. This is the case for instance of Media No Mad that also obtained from the crowd time and expertise (Larralde and Schwienbacher, 2010). Crowdsourcing differs in many ways from open-source practices (Brabham, 2008); some of these differences can be transposed to crowdfunding. An important distinction is that in the of opensource case, the idea belongs to the community who can then exploit it on an individual basis (there is no restriction on who can use it); in the case of crowdsourcing, the generated idea ultimately belongs to the company who will be the only one to exploit it. This distinction with open-source practices becomes even more obvious when related to crowdfunding, since capital cannot be shared. Unlike an idea or a software code, capital is not a public good in the economic sense that assumes non-rivalness and non-excludability (Samuelson, 1954). Under these conditions, a public good is a good that can be used by many consumers at the same time, without duplicating costs. Based on this discussion and in the spirit of Kleemann et al. (2008), we offer the following, refined definition: crowdfunding involves an open call, essentially through the Internet, for the provision of financial resources either in form of donation or in exchange for some form of reward and/or voting rights in order to support initiatives for specific purposes.
3. Related Literature Perhaps unsurprisingly, there is virtually no literature at all on crowdfunding. The little that exists can be primarily found in the one on crowdsourcing, which is a broader concept that encompasses crowdfunding. One of the very few academic articles on crowdfunding is from Kappel (2009) that distinguishes ex post facto (e.g., when a product is offered after financing is provided) from ex ante crowdfunding (e.g., financial support for lobbying or political activities). Wojciechowski (2009) discusses donations in connection with projects funded through crowdfunding. The author argues that social networks can become a worthwhile model of money collection for many charity organizations and NGOs. Whether it can be transposed to entrepreneurial activities is not discussed however. Relatedly, 6
Ghatak and Mueller (2009) develop a theoretical framework of labor donation theory to investigate under which conditions non-for-profit organizations can provide a better alternative to motivated workers than other forms of organizations. In the spirit of the contract failure literature (Glaeser and Shleifer, 2001), it is based on the view that limiting monetary incentives of owners attracts more easily donations, since it signals that the owners put a significant weight on the outcome and less on monetary gains. A related branch of research deals with bootstrap finance, which consists of using alternative financing ways than the traditional sources of external finance (e.g., bank loan, angel capital and venture capital). Several studies provide evidence of the different forms of alternatives used by bootstrapping entrepreneurs (see, Bhidé (1992), Winborg and Landstrom (2001) and Ebben and Johnson (2006), just to cite a few). Bhidé (1992) shows that even among the Inc. 500 companies in the US, most of them started by bootstrapping the company. Further financing methods for startups companies are analyzed, for instance, by Cosh et al. (2005), who examine a broader range of financing alternatives. Theoretical considerations about the optimal timing between using internal and external resources is provided by Schwienbacher (2007).
4. Data Collection and Variables To shed light on the structure of crowd-funded investments, we hand-collected data from various sources on all possible crowdfunding initiatives that we could identify on the Internet. Data collection took place end of 2009 and early 2010. Since there is no database available or even listing, we relied on the Internet to construct our sample. One advantage is that individuals using crowdfunding as a way to collect funds typically use the Internet to do so, as well as social networks, such as blogs, Facebook and Twitter. This helps identifying cases to construct our sample. Our focus was on crowdfunded ventures and projects, which largely excludes all initiatives made by artists. Our criteria to select entrepreneurs relying on crowdfunding have been done in two times. Initially, we made Internet research on crowdfunded ventures and projects that are explicitly associated with the word of “crowdfunding”. This way did not allow us to find sufficient number of such entrepreneurs. In a second time, we decided to revisit the definition of crowdfunding. That is, we selected all entrepreneurs whom use the Web 2.0 to generate funds for their investments via a large number of internet users. This step was essential because some entrepreneurs have recourse to it without knowing that their own fundraising is a kind of a “crowdfunding”. In total, we identified 7
88 cases, out of which we could collect sufficient (but still partially incomplete) information on 51 of them. Next to this, we sent a questionnaire, during the months of December 2009 and January 2010, to all the crowdfunded initiatives that we included in our sample in order to obtain further information on their motivations. For some, we could not identify a clear email address to contact them. In total, 69 entrepreneurs have been contacted and 21 completed questionnaires have been received (some only partially). The response rate in this survey is therefore around 30%. Despite the high response rate, the total sample remains relatively small; this in turn could inevitably raise potential statistical concerns. Indeed, this may induce some small-sample bias for which it is difficult to control; on the other hand, crowdfunding is a nascent phenomenon so that our initial sample of 88 initiatives converges toward the entire population. All the variables used in the analysis are defined in Table 1. Our criteria to include variables in our specifications are based on our research questions raised—namely the determinants of success by using crowdfunding. We count 7 groups of determinants: 1) On the funding outcome: Funds raised, Funds targeted, and Success (which is the ratio of the two former values); 2) Characteristics of the crowdfunding initiatives, including compensation for crowd-funders and type of investment (Donation, Active investment, and Passive investment); 3) Date of establishment and start of the crowdfunding initiative; this in turn allows to calculate the age of the organization at the time the crowdfunding initiative was started; 4) Country of registration/establishment; 5) Type of organizational forms used, namely as a Company, Specific project related, Non-profit organization, or as an Individual; 6) Communication methods used, where we consider the following types: Blogs, Own Internet site, CV of founder(s), LinkedIn, Twitter, MySpace, Facebook, and Other methods; 7) Type of industries.
5. Analysis on the Use of Crowdfunding 8
Table 2 provides results of the survey and summary statistics of the full sample. Results confirm that crowdfunding is a recent phenomenon. Indeed, over 80% of the respondents have used crowdfunding for projects or their own company most recently only (variable Recent CF); i.e., since 2007. 35.3% are from the United States and 49% from Europe. 63.2% are managed by a single founder, 15.8% by two founders and 21.1% by three founders (the highest number of founders observed in our sample). 70% of these founders hold a university degree, 10% are still attending the university. Raising money was a strong motivation for all respondents, getting public attention was relevant (or highly relevant) for over 85%, and obtaining feedback for the product/service offered was still relevant (or highly relevant) for about 60% of the respondents. Many of them combine crowdfunding with other sources of finance, notably with own money, friends & family money, business angel and government subsidy. 76.5% offer to their crowd-funders a reward, mostly in form of right to receive the own product (66.7% of the cases of these 76.5% of the sample) or shares that may yield dividends in the future (33.3%). Direct cash payment is expected in 22.2% of the cases where a reward/return is promised. Eventually, we note that, in 66.7% of the cases, other forms of reward are afforded; e.g., getting credit on an album or a film, by giving money to a charity of your choice, etc. Our study distinguishes different forms of investment: donation, active investment, and passive investment. As pointed before, pure donation constitutes 22% of crowdfunding. The rest represents investments (i.e., the crowd-funder expects to receive a return or reward), ventilated between active investment and passive investment. Both count respectively for 32% and 60%. In terms of means of communication, it is worthwhile that virtually all used very extensively the Internet as mode of communication with the “crowd”, evidencing the reliance on Web 2.0 for modern crowdfunding. Internet enables broad access to a community that may share similar goals and views. The most widely used methods of Internet is an own website, community blogs, Facebook and Twitter. Other methods are used by less than 50% of the respondents. However, only 20% of them (according to our survey) used a crowdfunding platform such as Couch Tycoon. Are entrepreneurs profit-oriented or not-for-profit-oriented? As discussed previously, most entrepreneurs have recourse to crowdfunding in connection with a specific project only (in 46% of the cases). This mean of funding is used by not-for-profit associations in 16% of the cases, whereas over 35% of our sample represents profit-oriented firms.
Several of these variables are correlated with each other, as evidenced in Table 3. While some are intuitive, others are worth being discussed. Interestingly, crowdfunding initiatives taking place as a company tend to involve more often active investments and fewer passive investments; company are more likely to enable individuals to provide input or vote on the project. One possible reason is that projects done outside a company may be smaller and simpler; also, interacting with the crowd for the any project requires an organizational structure that individuals may not possess. This correlation is also in line with the notion than investors may require more control than for other organizational forms; our simple correlation does however not offer any conclusive evidence on whether this is actually a main driver. Conversely, not-for-profit organizations less often offer active involvements of investors; this lends to think that their projects require little input from investors; also, conversely to companies, investors may put more trust into not-for-profit organizations; but again, this claim is highly speculative here. The variable Success positively correlates with the dummy variable Non-profit organization; this finding is confirmed in Section 6 in multivariate analysis and in line with theoretical arguments proposed, for instance, by Glaeser and Shleifer (2001) and Ghatak and Mueller (2009). Lastly, an interesting negative correlation exists between whether a reward is offered (variable Rewards is offered (Dummy)) and whether it is a passive investment (the variable Passive investment (Dummy)). This lends to expect that rewards and control are substitutes; investors may require more rewards if they cannot be involved in the happening of the initiative.
6. Determinants of Crowdfunding Success To understand what may affect crowdfunding outcomes, we use two different measures: one is simply the total amount raised through crowdfunding (the variable Funds raised, which is the natural logarithm of the total funds raised); the other one is the ratio of the total funds raised over the funds targeted (the variable Success). Table 4 shows multivariate regressions on the natural logarithm of Funds raised. Perhaps as expected, companies attract larger amounts, most likely because of their greater need for capital. However, more surprisingly, not-for-profit organizations tend to also raise more money. Entrepreneurial initiatives that yield a product tend to attract larger amounts of capital than those who offer a service. This result may be mechanical, as activities that yield a product will on average require large investments than for providing a service. Indeed, the former may require significant production facilities that lead to major capital expenditures upfront. A second possible reason for 10
this positive effect may stem from the fact that the investing “crowd” may be more tempted to provide money if they expect a tangible outcome; one reason could be that the provision of a product is contractible (Hart and Moore, 1988). In this case, they may favor initiatives that yield a product as opposed to a service. The analysis in Table 4 however abstracts from the actual financing needs, or targets. Therefore, we consider next possible determinants of success measured by the ratio of Funds raised over Funds targeted. Results are shown in Table 5. Not-for-profit organizations show up strongly as the primary driver of success, while other variables are never significant. However the result about not-for-profit organizations is very robust across specifications. The coefficient is also economically meaningful: compared to other forms, not-for-profits tend to raise 200% more funds than targeted through crowdfunding. This is remarkable and can be seen in line with the contract failure theory (Glaeser and Shleifer, 2001; Ghatak and Mueller, 2009) that these organizations are better at attracting outside funds because of their possible stronger focus on the outcome than on monetary gains. While we cannot exclude a possible bias due to the self-reporting of target amounts, any bias is likely to occur for all the initiatives; there is no specific reason to expect that entrepreneurs of not-forprofit organizations are prone to understate more than entrepreneurs of other organizational forms. In other words, such a bias would inflate values of the variables Success but it is likely to be similar across initiatives.
7. Discussion and Concluding Remarks To our knowledge, this is the very first empirical study directly dealing with crowdfunding on a larger sample. Existing studies, while providing useful insights into the process, are limited to individual case studies. While providing first-hand insights into the crowdfunding process, this study raises follow-up questions that should be examined in future research. For instance, are these investments worthwhile for individuals? Compared to other means of financings, crowdfunding opportunities exhibit several important differences that are likely to affect risk-return profile of investors and motivations for providing money to crowd-funders. From a general perspective, crowdfunding practices raise questions with respect to corporate governance and investor protection issues if most individuals only invested tiny amounts. Crowdfunders are most likely offered very little investor protection. This may lead to corporate governance issues, which in turn may turn into reputation concerns if some cases of fraud or bad governance are 11
uncovered. Crowd-funders have very little scope to intervene to protect their interests as stakeholders. Moreover, the fact that their investment is small is likely to create a lack of incentive to intervene. Therefore, trust-building is an essential ingredient for any successful crowdfunding initiative. It is therefore not a surprise that many of the observed crowdfunded initiatives are either projectbased or based on donations. In many cases, the financial return seems to be of secondary concern for those who provide funds. This suggests that crowdfunders care about social reputation and/or enjoy private benefits from participating in the success of the initiative (Glaeser and Shleifer, 2001; Ghatak and Mueller, 2009). This view is supported by our results, since not-for-profit organizations appear to be more successful in achieving their fundraising objectives. In any case, any entrepreneur will need to balance the pros and cons of different alternatives, since crowdfunding will unlikely be the least costly source of financing for most projects. First, the amounts received from each investor are small, generating potentially substantial transaction costs. And second, many entrepreneurs need to bring in also expertise, which most crowd-funders do not provide. However, a strong advantage of this form of financing is the attention that the entrepreneur may attract on his/her project or company. This can become a vital asset for many of them, especially for artists or entrepreneurs in need to present his talent and product to the “crowd” (as potential customers). In other cases, it is a unique way to validate original ideas in front of a specifically targeted audience. This may in turn provide insights into market potential of the product or service offered. In any case, crowdfunding may be viewed as a broader concept that purely raising funds: it is a way to develop corporate activities through the process of fundraising.
Bibliographical References Albors, J., J.C. Ramos and J.L. Hervas (2008), New Learning Network Paradigms: Communities of Objectives, Crowdsourcing, Wikis and Open Source. International Journal of Information Management 28, pp. 194-202. Bénabou, R., and J. Tirole (2006), Incentives and Prosocial Behavior. American Economic Review 96 (5), pp. 1652-1678. Bhidé, A. (1992), Bootstrap Finance: The Art of Start-ups. Harvard Business Review 70 (66), pp. 109117. Bosma, N.S., and J. Levie (2010), Global Entrepreneurship Monitor 2009 Executive Report, Babson Park, MA: Babson College, Santiago, Chile: Universidad del Desarollo, Háskólinn Reykjavík, Iceland: Reykjavík University and London, UK: London Business School. Brabham D.C. (2008), Crowdsourcing as a Model for Problem Solving: An Introduction and Cases. Convergence: The International Journal of Research into New Media Technologies 14 (1), pp. 7590. Cosh, A., D. Cumming and A. Hughes (2009), Outside Entrepreneurial Capital. Economic Journal 119, pp. 1494-1533. Ebben, J., and A. Johnson (2006), Bootstrapping in Small Firms: An Empirical Analysis of Change over Time. Journal of Business Venturing 21, pp. 851-865. Everett, C.R. (2008), Group Membership, Relationship Banking and Loan Default Risk: The Case of Online Social Lending. Available on SSRN: http://ssrn.com/abstract=1114428 Falk, A., and M. Kosfeld (2006), The Hidden Costs of Control. American Economic Review 96 (5), pp. 1611-1630. Freedman, S., and G.Z. Jin (2010), Learning by Doing with Asymmetric Information: Evidence from Prosper.com. Available on SSRN. Ghatak, M., and H. Mueller (2009), Thanks for Nothing? Not-for-Profits and Motivated Agents. LSE STICERD Research Paper Nbr. EOPP 014. Glaeser, E.L., and A. Shleifer (2001), Not-for-Profit Entrepreneurs. Journal of Public Economics 81, pp. 99-115. Hart, O., and J. Moore (1988), Incomplete Contracts and Renegotiation. Econometrica 56 (4), pp. 755 - 785. Howe, J. (2006) The Rise of Crowdsourcing. Wired http://www.wired.com/wired/archive/14.06/crowds.html 14 (6), available online at:
Howe, J. (2008), Crowdsourcing. Why the Power of the Crowd is Driving Future of Business. New York: Three Rivers Press. 13
Kappel, T. (2009), Ex Ante Crowdfunding and the Recording Industry: A Model for the U.S.?. Loyola of Los Angeles Entertainment Law Review 29 (3), pp. 375-385. Kleemann, F., G.G. Voß and K. Rieder (2008), Un(der)paid Innovators: The Commercial Utilization of Consumer Work through Crowdsourcing. Science, Technology & Innovation Studies 4 (1), pp. 5-26. Larralde, B., and A. Schwienbacher (2010), Crowdfunding of Small Entrepreneurial Ventures. Book chapter for “Entrepreneurial Finance” (Ed. D.J. Cumming), forthcoming at Oxford University Press. Lee, S.-H, D. DeWester, and S. R. Park (2008), Web 2.0 and Opportunities for Small Businesses. Service Business 2 (4), pp. 335-245. O’Reilly, T. (2007), What is Web 2.0: Design Patterns and Business Models for the Next Generation of Software. Communications & Strategies 65 (1), pp. 17-37. Samuelson, P.A. (1954), The Pure Theory of Public Expenditure, Review of Economics and Statistics 36, pp. 387-389. Schwienbacher, A. (2007), A Theoretical Analysis of Optimal Financing Strategies for Different Types of Capital-constrained Entrepreneurs. Journal of Business Venturing 22, pp. 753–781. Winborg, J., and H. Landstrom (2001), Financial Bootstrapping in Small Businesses: Examining Small Business Managers’ Resource Acquisition Behaviors. Journal of Business Venturing 16, pp. 235254. Wojciechowski, A. (2009), Models of Charity Donations and Project Funding in Social Networks. Lecture Notes in Computer Science 5872, pp. 454 – 463.
TABLE 1: Definition of Variables
Funding Outcome: Funds raised Funds targeted Success Crowdfunding Characteristics: (i) Donation (Dummy) (ii) Active investment (Dummy) (iii) Passive investment (Dummy) Reward is offered (Dummy)
Total funds raised (in Euro ) by the entrepreneur since the starting date of his crowdfunding process Total funds expected initially (in Euro ) from the crowdfunding process by entrepreneur Ratio of Funds raised over Funds targeted
Dummy = 1 if crowd-funders make only donation without any kind of financial return or reward, as understood in the Reward is offered variable Dummy = 1 if crowd-funders are involved in any way whatsoever in the venture or project they fund Dummy = 1 if crowd-funders are not involved in the venture or project they fund Dummy = 1 if entrepreneur gives any kind of return or reward to crowd-funder, such as direct cash payment, shares/stocks, including dividends in the future, own product, getting a credit on the album, the DVD, or the Film, etc. Dummy = 1 if the goal of the venture or project is the making of a product (conversely a service)
Outcome is a product (Dummy)
Date of Establishment and Start of Crowdfunding Initiative: CF date Year at which the crowdfunding process started CF age Recent CF (Dummy) Country of registration (Dummies): USA A European country Elsewhere Anglo-saxon countries (Dummy) Age in years of the crowdfunding process understood as the CF Date minus the establishment date Dummy = 1 if the year at which the crowdfunding process started in 2007 or later Dummy = 1 if the registered office of entrepreneur is located in USA Dummy = 1 if the registered office of entrepreneur is located in a European country Dummy = 1 if the registered office of entrepreneur is located elsewhere Dummy = 1 if the registered office of entrepreneur is located in an Anglo-Saxon country
Type of Organzational Form (Dummies): Company Dummy = 1 if the crowdfunding initiative is structured as a company Specific project related Dummy = 1 if entrepreneur raised money by crowdfunding in connection with a specific project only Non-profit organization Individual Dummy = 1 if entrepreneur is working on behalf of a non-profit-making association Dummy = 1 if the entrepreneur acts as an individual (e.g., freelance)
Communiation methods used (Dummies): Communiation methods used Dummy = 1 if the entrepreneur makes use of specific communication methods; we (Dummies) use a dummy variable for each of the following communication methods: (i) Blogs, (ii) Own internet site, (iii) CV of founder(s), (iv) LinkedIn, (v) Twitter, (vi) MySpace, (vii) Facebook, and (viii) Other methods. Nbr. of communiation methods used Whether social networks were used (Dummy) Number of communication methods listed above used in connection with the crowdfunding initiative; it is the sum of all the different dummy variables above under Communication methods used. Dummy =1 if at least one of the following communication methods is used: Facebook, Twitter, blogs, LinkedIn, MySpace; these methods are characterized by facilitating social networking.
Type of Industries (Dummies): Industry
Dummy = 1 if the entrepreneur is in a specific industry; we use a dummy variable for each of the following industries: (i) Education, (ii) Film/Music, (iii) Finance, (iv) Food/Restaurant, (v) ICT, (vi) Journalism, (vii) Medical, (xiii) Politics, (ix) Recycling, (x) Sport, and (xi) Tourism.
To convert amount from other currency in euro, we computed an annual average rate (from 1/20/2009 to 1/20/2010) of the euro foreign exchange reference rates from the European Central Bank online statistics (available at: www.ecb.int).
TABLE 2: Characteristics of Crowdfunded Projects -- Summary Statistics and Survey Output
Panel A: Summary Statistics Variables Mean Std. Dev. Median Min Max Nbr. Obs.
Funding Outcome: Funds raised Funds Targeted Success 3.5 million 11.0 million 245.654 15.0 million 27.7 million 664.375 28,583 107,185 2.500 36 60 0.222 82.1 million 82.1 million 2873.937 33 39 33
Crowdfunding Characteristics: (i) (ii) (iii) Donation (Dummy) Active investment (Dummy) Passive investment (Dummy) 0.220 0.320 0.600 0.860 0.529 0.418 0.471 0.495 0.351 0.504 0.000 0.000 1.000 1.000 1 0 0 0 0 0 1 1 1 1 1 50 50 50 50 51
Reward is offered (Dummy) Outcome is a product (Dummy)
Date of Establishment and Start of Crowdfunding Initiative: CF date CF age Recent CF (Dummy) 2007.6 1.667 0.806 2.048 2.523 0.401 2008 1.000 1.000 2001 0 0 2009 9 1 36 30 36
Country of registration (Dummies): (i) (ii) (iii) USA A European country Elsewhere 0.353 0.490 0.157 0.588 0.483 0.505 0.367 0.497 0.000 0.000 0.000 1.000 0 0 0 0 1 1 1 1 51 51 51 51
Anglo-saxon countries (Dummy)
Type of Organzational Form (Dummies): (i) (ii) (iii) (iv) Company Specific project related Non-profit organization Individual 0.360 0.460 0.160 0.020 0.485 0.503 0.370 0.141 0.000 0.000 0.000 0.000 0 0 0 0 1 1 1 1 50 50 50 50
Communiation methods used (Dummies): (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) Used blogs Used own Internet site Used the CV of founder(s) Used LinkedIn Used Twitter Used MySpace Used Facebook Used other methods 0.689 0.860 0.106 0.111 0.467 0.156 0.533 0.48 3.956 0.804 0.468 0.351 0.312 0.318 0.505 0.367 0.505 0.51 1.954 0.401 1.000 1.000 0.000 0.000 0.000 0.000 1.000 0.000 4.000 1.000 0 0 0 0 0 0 0 0 1 0 1 1 1 1 1 1 1 1 8 1 45 50 47 45 45 45 45 46 45 50
Nbr. of communiation methods used Whether social networks were used (Dummy)
Type of Industries (Dummies): (i) (ii) (iii) (iv) (v) (vi) (vii) Education Film/Music Finance Food/Restaurant ICT Journalism Medical 0.020 0.216 0.098 0.098 0.137 0.157 0.020 0.140 0.415 0.300 0.300 0.348 0.367 0.140 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 51 51 51 51 51 51 51
(viii) (ix) (x) (xi)
Politics Recycling Sport Tourism
0.020 0.059 0.118 0.059
0.140 0.238 0.325 0.238
0 0 0 0
0 0 0 0
1 1 1 1
51 51 51 51
Panel B: Additional Statistics Based on Survey Output Questions Number of founders: One founder only: Two founders exactly: Three founders exactly: Question: "Do founders hold a university degree?" Yes 70.0% No 20.0% Still attending university 10.0% [Note: if more than one founder involved, we consider each founder separately.] 19 19 19 63.2% 15.8% 21.1% 19 19 19 Answers (%) N
Question: "Do people (the “crowd”) who invest in your company/project expect to receive return or reward from their investment?" Yes No, they only make a donation Sub-question: "If yes, what kind?" Direct cash payment (other than dividends from shares) Shares/stock, including dividends in the future Right to receive own product Other 22.2% 33.3% 66.7% 66.7% 9 9 9 9 76.5% 23.5% 17 17
Question: "If you give investors shares, do you allocate voting rights to them?" Yes No 18.2% 81.8% 11 11
Question: "If other sources of finance than crowdfunding are used, please specify which one(s)." Bank loan Contributions from family and/or friends Business angel Founder's own money Government subsidy Other 0.0% 18.8% 18.8% 25.0% 18.8% 43.8% 9 9 9 9 9 9
Question: "Do you make use of a crowdfunding platform (e.g. Couch Tycoon)?" Yes No 20.0% 80.0% 15 15
Question: "What constitutes your main motivations for using crowdfunding?" (Nbr. Obs. = 14 for all the three motivations provided to respondents and listed below.) High relevant Raise money Getting public attention for my company/project Validate my product/service before selling it (market survey) [Respondents could also cite other motivations; they are not listed here explicitly.] 92.9% 64.3% 35.7% Relevant 7.1% 21.4% 21.4% Neutral 0.0% 7.1% 7.1% Somewhat relevant 0.0% 0.0% 0.0% Not relevant at all 0.0% 7.1% 35.7%
Note: All the variables in Panel A are defined in Table 1. Information shown in Panel B was collected through the survey.
TABLE 3: Correlation Matrix
(1) (1) Company (Dummy) (2) Non-profit organization (Dummy) (3) Outcome is a product (Dummy) (4) Reward is offered (Dummy) (5) Passive investment (Dummy) (6) Active investment (Dummy) (7) Success (8) Nbr. communication methods used (9) Whether social networks were used (Dummy) (10) CF age (11) Recent CF (Dummy) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)
1.0000 -0.3273** 0.1368 0.1901 -0.3146** 0.4124*** -0.0101 0.0895 0.1667 0.1520 0.0624 1.0000 -0.0175 -0.6086*** -0.1949 -0.2934** 0.3159* -0.1263 -0.0546 -0.0529 -0.2956** 1.0000 -0.0415 -0.0490 -0.0275 0.2458 0.1630 0.1280 0.1881 -0.0336 1.0000 0.3765*** 0.2768* 0.0938 0.2804* 0.2306 -0.0240 0.0033 1.0000 -0.4901*** 0.0056 0.2543* 0.2041 0.0723 0.0235 1.0000 -0.0728 -0.1423 0.0214 -0.3443* -0.0939 1.0000 -0.2787 0.0177 -0.0596 -0.3672** 1.0000 0.6241*** 0.0379 0.0219 1.0000 -0.1114 -0.0535 1.0000 0.2753 1.0000
Note: All the variables are defined in Table 1. * Significance levels for 10%. ** Significance levels for 5%. *** Significance levels for 1%.
TABLE 4: Regression Analysis on the Amount of Funds Raised
Variales Non-profit organization Company Outcome is a product Reward is offered Passive investment Active investment (1) 2.92 * 5.20 *** 3.47 *** 0.25 (2) 1.66 5.53 *** 4.29 *** 1.26 -0.79 3.92 *** (3) 2.57 * 4.80 *** 3.28 *** (4) 3.88 *** 3.25 *** 3.17 ***
Whether social networks are used Number of communication methods used CF age
-1.63 -0.36 -0.09 -0.39
F-Stat (P-Value) Adj R-2 Nbr. Obs.
0.002 58.30% 23
0.001 61.60% 23
0.001 59.40% 23
0 80.40% 23
Note: The dependent variable in all the regressions is the natural logarithm of Funds raised. All the variables are defined in Table 1. The method of estimation is OLS. A constant term is included in all the regressions, whose coefficient is not reported. Coefficient tests are based on ordinary standard errors. Significance levels: *** for 1%, ** for 5%, and * for 10%.
TABLE 5: Regression Analysis on Success of the Crowdfunding Process
Variables Non-profit organization Company Outcome is a product Reward is offered Passive investment Active investment (1) 2.07 *** 0.15 0.12 -0.1 0.57 0.95 0.49 -0.19 (2) 2.00 *** (3) 2.17 *** 0.36 0.25 (4) 1.98 ** 0.24 0.17
Whether social networks are used Number of communication methods used CF age
0.19 -0.25 -0.04 0.02
F-Stat (P-Value) Adj R-2 Nbr. Obs.
0.125 25.60% 20
0.023 42.00% 20
0.082 31.60% 20
0.119 26.30% 20
Note: The dependent variable in all the regressions is the ratio of Funds raised over Funds targeted. All the variables are defined in Table 1. The method of estimation is OLS. A constant term is included in all the regressions, whose coefficient is not reported. Coefficient tests are based on ordinary standard errors. Significance levels: *** for 1%, ** for 5%, and * for 10%.