Female-led startup companies do not perform as well as their male-led counterparts in part because of gender discrepancies among venture capitalists, according to recent University research.
This gap in performance, the research suggests, is due to the majority male-driven nature of venture financing syndicates.
The study, conducted by Business Ph.D. candidate Sahil Raina, found there was a 25 to 30 percent difference in the average performance rates of male-led startups and female-led startups, with male-led startups more successful overall. Findings indicated that the difference in rates depended on whether the companies were initially financed by venture capital partners with at least one female general partner in the financing syndicate, or if the syndicate was all male.
Raina said his interest in this research began after reading that female entrepreneurs in Silicon Valley were having trouble working with venture capitalists. However, because the stories he read were lacking empirical evidence, Raina decided to study the issue himself.
Venture capital financing works by offering money to startup companies in rounds until the company either goes public or is bought by another company. Companies pitch their startups to firms and work toward gaining their own funding during the process. At that point, the venture capitalists have the chance to exit from financing the startup.
“It started off as this investigation into what’s going on in terms of gender and venture capital finance,” Raina said. “Venture capital financing provides money to early stage startups — like high-tech startups that aren’t public.”
Raina used CrunchBase, a database for the high-tech sector, to collect information on over 10,000 venture capital financing rounds and about 3,600 startups.
CrunchBase included the names of the entrepreneurs who founded the startups as well as the names of the general partners who were leading the venture capitals, allowing Raina to use the names to assign gender to all of the people in the sample.
For the names Raina said he could not immediately assign a gender to — such as gender-ambiguous names or those that were uncommon American names — he used online resources to try to determine if they were male or female. Names that weren’t distinguishable were excluded from the study.
According to Raina, about 10 percent of the entrepreneurs in the sample were women, and only about 15 percent of the startups had just one female founder. Overall, in entrepreneurship in the United States, about 36 percent of small businesses have majority female ownership, Raina said.
Raina said he wanted to explore specifically whether the performance of the startups differed by either the gender of startup’s founder or the venture capitalist funding it. For the purpose of the study, success was determined by whether venture capitalists exited a project, signalling it was either public or had been bought by another company.
“For startups that had only male founders, I found they had an exit rate of about 27 percent,” Raina said. “But for startups that had at least one woman founder, the exit rate was more like 17 percent.”
Considering the 10 percentage point difference, Raina then examined whether venture capital financing or advising might play a role in the gender gap.
He looked at rounds that had at least one female general partner in the financing syndicate versus financing rounds with all male partners in the syndicate, and compared the startups that were financed in these two types of financing rounds.
Startups that were financed by either of these two groups had similar proportions of their portfolios led by women founders, according to the study. In both venture capital groups, about 16.5 percent of the portfolios were startups led by women founders.
For male-led startups, there was no difference in performance based on whether they were financed by a syndicate with all male partners or by a syndicate with at least one female partner. On average, male-led startups successfully exited at 21 to 22 percent of the time.
Among syndicates that had at least one female general partner, there was no difference in the average performance rates of male- led startups and female-led startups they funded. However, for financing syndicates that had all male general partners, there was a 25 to 30 percent difference in the average performance rates of male and female-led startups.
“All of this difference is coming from the female-led startups doing much better when they are initially financed by a syndicate run by at least one female GP,” Raina said.
Raina said this makes him believe that the venture capitalists with female general partners are able to better select female-led startups, and this implies that some valuable startups might not succeed depending on who is financing them.
Finance Prof. Sugato Bhattacharyya, chair of Raina’s dissertation committee, said the most intriguing part of Raina’s research was that venture capitalist groups with women as partners effectively selected firms with women better.
“What it comes down to eventually is the notion that when people look at women’s participation in whatever way in this sub-area, one culprit that held up is the widely known demographic fact, at least in North America, women tend to go less into STEM fields historically,” he said. “That in itself might be causing disparity.”
Given the results, Raina said along with emphasis on STEM — science, technology, engineering and math majors — it may be beneficial to encourage women to go into venture capitalism or finance-related majors as well. He said encouraging women to work in venture capital will result in higher participation and success for female entrepreneurs.
Bhattacharyya noted, however, that because the research is ongoing, it is difficult to say definitively whether a disparity among startups exists.
“These things are never static,” Bhattacharyya said. “They evolve. By the time we find that in 1996 a big disparity existed and then in 2015, is it relevant or is it not? By this time there will be more women capitalists, more women entrepreneurs, and maybe it doesn’t exist anymore.