To illustrate the old adage that birds of a feather flock together, there may be no better example than the venture capital industry.
A recent study finds that venture capitalists have a strong tendency to team up with other VCs whose ethnic and educational backgrounds are similar to their own. Unfortunately, that tendency turns out to be bad for business.
“At the early stage of a company, you want the people around the table to challenge each other.”
In the paper The Cost of Friendship, three Harvard researchers show that the more affinity there is between two VCs who co-invest in a new company, the less likely it is that the company will succeed.
"Much of the homophily literature in business research talks about the positive benefits of working with people who are similar to you—ease of communication, comfort level, and the like," says Paul Gompers, the Eugene Holman Professor of Business Administration at Harvard Business School, who cowrote the paper with HBS Associate Professor Yuhai Xuan and Vladimir Mukharlyamov, a graduate student in the Economics department at Harvard. "What we show is that, in this context, the effects can be quite negative."
The team set out to answer a few key questions: What specific characteristics influence individuals' desire to work together on an investment deal? And given that influence, how does affinity affect investment performance? Do common characteristics lead to better communication, which then leads to better decisions? Or does like-mindedness lead to narrow decision-making, to the detriment of the deal?
The research began with a database of 3,510 individual venture capitalists and their investments in 12,577 companies between 1973 and 2003. Over the course of six years, the research team collected detailed biographical information on each VC, including ethnicity, educational background, and employment history. They then looked at who had invested with whom, and what those co-investors had in common.
Across the board, the researchers found that venture capitalists tended to co-invest in deals with other VCs who possessed similar characteristics. This was true regardless of whether the similarities were ability-based or affinity-based. For example, two VCs who graduated from the same undergraduate school were 34.4 percent more likely to collaborate on a deal than were two VCs from different alma maters. And the probability of collaboration between VCs increased by 39.2 percent if they were members of the same ethnic minority group.
The data held up with what Gompers had observed qualitatively in his two decades of studying the venture capital industry. "There are strong affinity groups with Indian venture capitalists and entrepreneurs and with Chinese venture capitalists and entrepreneurs," Gompers says. "And there's sort of a cabal of Jewish entrepreneurs and VCs as well."
The team then examined how these similarities had affected the outcomes of the portfolio companies in the study. (For the purposes of the paper, a successful outcome was defined as one in which a company eventually filed for an initial public offering.)
They found that the probability of success decreased by 17 percent if two co-investors had previously worked at the same company—even if they hadn't worked there at the same time. In cases where investors had attended the same undergraduate school, the success rate dropped by 19 percent. And, overall, investors who were members of the same ethnic minority were 20 percent less successful than investors with different ethnic backgrounds.
It dawned on the researchers that affinity might make it easier for one venture capitalist to guilt-trip another into making a bad deal—doing a favor for a friend. "We thought it could be that they only syndicate the deals to their friends that they can't get anyone else to do," Xuan says.
To test for that possibility, the team assessed the 12,577 investments according to measures that had proven to be indicators of future success, according to previous research. Such indicators included whether a company's founder had a history of founding successful companies, the stage of the portfolio company (risky early stage versus less-risky later stage), and how much media attention the company had received at the time of investment.
Controlling for these factors, they found that the quality of the deals was not apparently affected by co-investor affinity. In other words, birds of a feather did not necessarily pick worse investments than birds of different feathers on day one. "It's not like we invest into a deal that's bad to start with, and therefore we get a bad outcome in the end," Xuan says.
Rather, the lack of success among similar investors seemed to lie in the decisions that followed the investment.
In addition to granting cash, venture capitalists are heavily involved in hiring or firing the CEO of the portfolio company, choosing a board of directors, devising an overall strategy, identifying potential partners, and so on. Indeed, the researchers found that the negative affinity effect was strongest in early-stage deals, which generally require more input from investors than do later-stage deals.
"[The] lower likelihood of success of co-investments between venture capitalists that share similar characteristics is triggered by them making inefficient decisions or even mistakes that they would otherwise avoid," the researchers write in The Cost of Friendship.
They attribute this inefficiency to "groupthink," the psychological phenomenon in which members of a group make poor decisions because they fail to consider viewpoints other than their own. "When you are really familiar with each other, you tend not to go outside of your circle to get an outside opinion," Xuan says.
The findings are in line with some organizational behavior studies, which have found that that work groups perform better when members learn from one another's disparate experiences. "I think this carries over to venture-funded startups, in which having a diversity of venture capitalists around the table is actually critical to their success," Gompers says. "Take two people who once worked at Google, who went to Harvard Business School, and who are Indian American. They probably look at things in a very similar way and are unlikely to challenge each other. But at the early stage of a company, you want the people around the table to challenge each other."
Gompers and Xuan make a point of sharing the finds with students in the MBA program at HBS, many of whom pursue careers in the venture capital industry. In fact, people with Harvard MBAs make up 24.4 percent of the professional ranks at venture capital firms in the United States, according to a study by PitchBook. A network that powerful must beware the power of groupthink and collaborate with other networks, the professors advise.
"Students come to HBS because, in addition to having access to great faculty like Yuhai, they get the opportunity to interact with other students who are extremely talented and successful," Gompers says. "And of course they should continue to tap this network.
"But it's likely that if you're an HBS MBA, you think like other HBS MBAs, because you took the same courses from the same professors. And it's important for students to realize that it might be useful to have a diversity of people around the table when you make investment decisions or you're working on new ventures. That, at least for me, is an important prescriptive element of the paper."