Angel Financing and the Performance of High‐Tech Start‐Ups

Abstract

In this paper, we investigate what drives the performance of high‐tech start‐ups receiving angel financing, while taking a closer look at the capabilities (i.e., experience) and investment behavior of business angels (BAs). We exploit a new data set (extracted from Crunchbase), which consists of 1,933 high‐tech start‐ups that received at least one financing round from a BA. The results indicate that the experience of BAs in early stage investments is positively associated with additional receipt of follow‐on rounds of financing and sequential capital injections from venture capitalists (VCs). Later‐stage experience is positively associated with the start‐up's success (i.e., probability to be listed or acquired), but reduces the need for new VCs to invest in the start‐up. Furthermore, we find consistent evidence that start‐ups that combine BA and VC financing experience higher levels of funding amounts, additional VC financing, and an improved likelihood of success. Finally, we find that the co‐localization of BA investors and start‐ups in the same area facilitates the attraction of VC financing.

Notes

1 According to Shane (2012), funding from family, friends and fools (FFFs) should not be considered part of the informal capital market. However, it is worth noting that FFFs far outweighs funding from BAs and VCs as reported by the Angel Capital Education Foundation (2010). Start‐up funding from VCs and angel investors totals approximately $20.8 billion annually, while FFFs contribute nearly three times the amount of capital, with approximately $60 billion. An astonishing 87 percent of all funding of private companies in the United States comes from FFFs as opposed to other means (Angel Capital Education Foundation, 2010).

2 However, this market is difficult to accurately quantify because of its informal nature (Fenn and Liang 1998; Prowse 1998). Available figures mainly refer to the U.S. market and generally suggest that BA financing dominates VC financing, both in terms of number of invested firms and of total amount of financial investment (Sohl 2005; Wiltbank 2005; Wiltbank et al. 2009). According to Crunchbase, the U.S. angel market grew at an annual rate of 33 percent between 2007 and 2013 (Hellmann and Thiele 2015). Fenn and Liang (1998) also reported that in the United States, for every one firm that raised VC, six raised a BA investment and that approximately one‐third of firms that went public were funded by VCs and two‐thirds by BAs. Moreover, the number of individuals who fulfil all the conditions for becoming BAs but that have never invested has been estimated to be 850,000 in Europe and 1.75 million in the United States (CVR 2003).

3 http://www.crunchbase.com.

4 Hellmann et al. (2013) suggest dynamic substitutes patterns between BAs and VCs, which constitute alternative investment opportunities that do not mix well together. They find that deals that originally raised BA capital are less likely to then obtain VC funding and, in case of funding, raise lower amounts. The effect is more pronounced for single‐company BAs than for multiple‐company BAs, or for those that invest together through an angel fund. Hellmann and Thiele (2015) develop a theoretical model of how BAs and VCs interact. VCs and BAs are “friends” because they rely upon each other's investments, but they are also “foes” because VCs no longer need the angels when later stages are reached.

5 CrunchBase is operated by TechCrunch, one of the most influential technology blogs in the United States (http://techcrunch.com). The data set is quite new and it shows a good potential for research purposes. The data set can be downloaded at http://info.crunchbase.com/about/crunchbase-data-exports.

6 http://techcrunch.com/2013/07/12/crunchbase‐and‐angellist‐have‐a‐partnership.

7 In our sample the proportion of BA‐backed start‐ups that received VC is 56 percent (1,086/1,933). Madill et al. (2005) find that the proportion of BA‐backed firms in Canada that received both BA and VC financing is 57 percent. In Goldfarb et al.'s (2013) study on California‐based firms, this proportion is 78 percent.

8 Our measure of BA experience is constructed from the database itself. This is a clear limitation that we acknowledge. However, the use of prior investments as a proxy for investment experience is quite common in the entrepreneurial finance literature (see, e.g., Sørensen 2007, in the context of venture capital).

9 To exclude the possibility of M&A exits that possibly disguise unsuccessful investments sold to the management for a nominal sum (liquidation sales), we also checked the acquisition price for the sub‐sample of companies for which such information was available (109 firms). The average acquisition price is $166 million and the median acquisition price is $50 million. The 5th percentile of the distribution of the acquisition price is $2.6 million, while the 95th percentile is $930 million.

10 It is worth pointing out that the dependent variable is defined only if the focal company receives a subsequent round of financing.

Additional information

Notes on contributors

Annalisa Croce

Annalisa Croce is associate professor in the Department of Management, Economics and Industrial Engineering at Politecnico di Milano.

Massimiliano Guerini

Massimiliano Guerini is assistant professor in the Department of Management, Economics and Industrial Engineering at Politecnico di Milano.

Elisa Ughetto

Elisa Ughetto is assistant professor in the Department of Management and Production Engineering at Politecnico di Torino.