Human capital signals and entrepreneurs’ success in equity crowdfunding

Interestingly, the number of solo entrepreneurs who search for equity finance on the Internet is not negligible. For instance, in the sample we use in the present study, more than 10% of the entrepreneurs run their start-ups alone.

SiamoSoci was the first equity crowdfunding platform created in Italy. More than 2 years after its establishment, the Consob Regulation no. 18592 “The collection of risk capital on the part of innovative start-ups via on-line portals” (http://​www.​consob.​it/​mainen/​documenti/​english/​laws/​reg18592e.​htm) entered into force. This regulation included a set of rules and requisites for the managers of equity crowdfunding platforms aimed at reducing operational and legal risks and the risk of litigation and fraud. The regulation also established a register where platforms had to be included to be allowed to use the label “equity crowdfunding platforms”. In light of the new Italian regulation, SiamoSoci could not use the label equity crowdfunding platform any more and in 2014, it started changing its activity to become a marketplace for entrepreneurs searching for business angels’ investments. However, despite the lack of formal recognition by Consob, during the period under scrutiny in this study, SiamoSoci functioned de facto as an equity crowdfunding platform: it matched entrepreneurs and investors through the Internet, although the finalization of the financing contract had to be done off line. As a detailed analysis of this latter legal aspect goes beyond the purposes of this study, we refer to SiamoSoci as an equity crowdfunding platform.

During the period of observation, 187 campaigns were launched on SiamoSoci, but 21 of them (about 11%) were still open and raising capital when we stopped the data collection.

For about half of sample entrepreneurs, the number of LinkedIn connections was collected only at mid-December 2013. As for these individuals, the number of connections when the campaign was launched was unknown, we could have not computed ln_LinkedIn_connections. Hence, we resorted to regression imputation. Specifically, we first estimated an OLS regression model on the sole individuals for which the number of LinkedIn connections was available both when the campaign was launched and at mid-December 2013. The dependent variable was the number of LinkedIn connections when the campaign was launched, while the independent variables were the number of connections at mid-December 2013, the number of months elapsed since campaign launch and the number of entrepreneurs involved in the start-up. The estimated coefficients were then used to impute the missing values.