4. The Crowd–Entrepreneur Relationship in Start-Up Financing

See, for example, “Facebook to Buy Virtual Reality Firm Oculus for USD 2 Billion” by Reed Albergotti and Ian Sherr, Wall Street Journal, March 25, 2014, “Facebook Plays Platform Catch-Up with USD 2 Billion Oculus Deal” by Hannah Kuchler and Tim Bradshaw, Financial Times, March 26, 2014, or “Facebook Buys VR Startup Oculus for USD 2 Billion” by Cade Metz, Wired, March 25, 2014.

“When Crowdfunding Goes Corporate: Kickstarter Backers Vent over Facebook’s Oculus Buy” by Victor Luckerson, Time, March 26, 2014.

These numbers of the crowdfunding market are sourced from the Massolution Crowdfunding Industry Report (2015), available at: http://​reports.​crowdsourcing.​org/​index.​php?​route=​product/​product&​product_​id=​54 (last consulted on July 12, 2016).

This section partly draws from Belleflamme and Lambert (2014).

Cumming et al. (2015) consider different types of crowdfunding models (i.e. Keep-It-All vs. All-Or-Nothing) and show that they allow to allocate the risk differently between the crowd and the entrepreneur. Without loss of generality, the discussion in Belleflamme et al. (2014) does not account for this variation of crowdfunding models.

We have here a form of behavior-based price discrimination as consumers self-select into one group and are then charged a specific price corresponding to their choice; see Fudenberg and Miguel Villas-Boas (2006) for a general analysis of behavior-based price discrimination and Belleflamme and Peitz (2015) for a textbook treatment.

See also Sahm et al. (2014), who slightly correct the analysis on profit sharing and, thereby, simplify it.

Ellman and Hurkens (2015) also examine a crowdfunding model that allows the crowd to pre-order the product and in which entrepreneurs can commit to produce only if aggregate funding exceeds a defined threshold. Yet, in their model, pre-ordering does not confer any additional community benefit. Their objective is to determine the optimal crowdfunding mechanism in the presence of two conflicting forces: a high threshold allows the entrepreneur to set higher prices for high-type buyers, while a low threshold raises the probability of production. Kumar et al. (2016) also model threshold choice, but they consider a continuum of consumers.

See Chemla and Tinn (2016) whose model emphasizes the importance of learning about demand as an essential reason why entrepreneurs engage in the pre-ordering mechanism.

Community benefits do not have to be confused with the rewards that are often offered by entrepreneurs to the participating crowd. Rewards simply represent, in both business models, a materialization of these community benefits.

See https://​www.​kickstarter.​com/​projects/​597507018/​pebble-e-paper-watch-for-iphone-and-android/​posts/​273665 (last consulted on July 11, 2016).

This quote is also stressed by Agrawal et al. (2014). See https://​blog.​getpebble.​com/​2012/​05/​08/​and-one-more-thing/​ (last consulted on July 11, 2016).

See www.​crowdcube.​com and www.​kickstarter.​com/​help/​stats (last consulted on July 11, 2016).

In most cases the products/rewards promised by the entrepreneur are delivered (see Mollick 2014) in part because a failed delivery history would also establish a negative reputation of the entrepreneur which decreases funding success when returning to raise money on Kickstarter (Li and Martin 2016).

According to the Small Business Administration (SBA), only about 50% of businesses survive five years or longer. See https://​www.​sba.​gov/​sites/​default/​files/​advocacy/​SB-FAQ-2016_​WEB.​pdf (last consulted on November 28, 2016).

Such an exit event might only occur several years after the original investment, if at all. Hornuf and Schmitt (2016) report that until the end of 2015 only seven start-ups that equity crowdfunded in Germany offered exit opportunities to their investors.

The problem of adverse selection, introduced by Akerlof (1970), can be illustrated with a simple example. Assume the venture is of either a low-quality (worth USD 100) or a high-quality type (worth USD 200). Investors know that there is an equal chance that they are dealing with a low-quality or high-quality venture but cannot distinguish between the two. In that case, backers would rationally price the venture at the average value of (50%*USD 100+50%* USD 200=) USD 150. This implies an undervaluation of USD 50 for the high-quality venture and an overvaluation of USD 50 for the low-quality venture. As a result, low-quality ventures would quickly start to flood the market for early stage finance, which in the extreme case could lead to a market breakdown where no venture successfully gets financed.

Portals are of course not the only actors capable of mitigating such information asymmetries. We refer to Ahlers et al. (2015), Ralcheva and Roosenboom (2016) and Vismara (2016), among many others, whose work focuses on entrepreneurs and stresses the role of certification and quality signals.

Traditionally, both business angels and venture capitalists would use their personal connections and networks, and attend local start-up events or dedicated start-up conferences as means to access new potential investment opportunities. The typical way to source vital information would be through various documents and/or face-to-face communication.

Hildebrand et al. (2016) consider incentive effects for group leaders on the crowdlending portal Prosper. Their results suggest that similarly to crowdinvesting, group leaders on Prosper much more carefully screen and choose the listings to be funded when they have sufficient “skin in the game.”