Are Venture Capitalists Hopelessly Biased, or Deceptively Rational?

By Bala Vissa

Close scrutiny of VC deal terms in India reveals a nuanced pattern of objectivity and discrimination.

No matter how rational we may claim or want to be, no human being is entirely free of bias. The challenge is to be aware of one’s own bias – or, failing that, to design processes that mitigate bias – so that objective considerations can prevail. This challenge becomes especially pertinent in a field such as venture capital, where intrinsic uncertainty renders decision makers particularly prone to irrational influences.

Some recent research suggests that bias is rife in the world of venture capitalists. Studies using U.S. data have found, for example, that VCs show a marked preference for founders who share their ethnicity, regardless of financial performance. The same tendency towards homophily (favouring socially similar others) may also help explain the male-dominated sector’s stubborn, sizeable gender gap.

It would be premature, however, to count out rationality just yet. In a recent paper forthcoming in Organization Science, we delved into how social similarity influences pricing in VC deals. Our findings suggest that VCs are reasonably adept at structuring deals that reflect both the benefits and the costs of social similarity. Intriguingly, this sometimes involves VCs providing more favourable terms to founders who are socially dissimilar to themselves. 

India’s VC market

Our study diverges in several key ways from past research on how social similarity regulates VCs’ deal making. First, instead of using data from the United States, we looked at 622 deals within India’s rapidly maturing VC industry. Second, in addition to studying how VCs select companies to fund, the focus of much prior research, we examined how VCs price the capital they provide – specifically, valuation and downside risk protection. The latter refers to contractual terms such as liquidation preferences, cumulative dividend rights, redemption rights and anti-dilution rights, which regulate VCs’ future cash flows from the deal. Third, we factored in similarity between VCs and founders on two under-studied social attributes: the Indian region and caste that VCs and founders belong to.

Although it is largely viewed today as an unwanted vestige of an imperfect past, caste stratification is a lingering influence in Indian society, especially in rural areas where recent social and technological developments are not as forcefully evident. Conversely, while regional differences in India can include wide linguistic and cultural gaps, they do not carry caste’s automatic associations with social hierarchy.

As with past studies, we also measured the VCs’ financial returns from each deal. Importantly, however, we used internal rate of return (IRR) as our metric, rather than indicators such as whether there was a successful IPO launch. We reasoned that IRR was a better barometer of value creation driven, in part, by the collaboration between founders and investors.

Social similarity and dissimilarity

Our analyses yielded a nuanced picture on how social similarity matters. When it came to regional similarity, we found that VCs confer higher pre-money valuations to companies led by regionally similar founders. This pattern is consistent with a logic whereby VCs rationally anticipate the benefits of social similarity – it is easier to build a collaborative, trusting relationship with similar others – and price it in at the deal-making stage. Yet, strikingly, VCs also engaged in greater downside risk protection on these very same deals!

In essence, VCs seem to recognise the dangers of social similarity (such as potential free riding or lack of optimal effort by regionally similar founders) and attempt to protect themselves through more stringent downside protection terms. This pattern is consistent with rational behaviour by VCs rather than evidence of bias. Moreover, we found that the strategy seems to be working, as greater regional similarity was indeed correlated with higher financial returns for the VCs.

Switching focus to caste stratification, we found no statistically significant, overarching impact on deal flow. However, matches between higher-caste VCs and lower-caste founders were generally associated with higher valuations as well as better financial performance for the VC. This interaction was especially prominent when the lower-caste founders had elite educational credentials, such as graduating from the prestigious IITs and IIMs. Again, this is consistent with rational behaviour by VCs rather than non-rational motivations, as would be the case if high-caste VCs were indiscriminately funding low-caste entrepreneurs out of altruistic notions of noblesse oblige.

Rank-ordered social attributes

Our conclusions point to a broader logic that may govern the rational uses of homophily in the VC space. As stated above, the primary difference between regional similarity and caste similarity is that the latter is rank-ordered, i.e. associated with social hierarchy. Our findings suggest that VCs are quite rational in anticipating and pricing-in the benefits as well as the pitfalls of social similarity when the social attribute in question is not rank-ordered – such as regional similarity in India. We would expect similar results in parallel contexts, e.g. how similarity or difference in nationality affects the VC market in Western Europe.

In contrast, the more a social attribute is infused with hierarchical considerations, the more nuanced its effects on pricing in VC markets, with dissimilarity-based mechanisms driving behaviour. While the caste system is unique to India, rank-ordered attributes can be found in every culture. For example, caste could be likened to immigration or nativity status in the developed Western economies. While VC as a global industry has a long way to go as regards diversity, the good news is that investors seem perhaps less biased than some prior research suggests. It may also help to highlight that the bit of extra effort required to find outstanding talent among marginalised groups can pay off just as much as close relationships carefully cultivated with socially similar founders.

Bala Vissa is a Professor of Entrepreneurship at INSEAD.

Kim Claes is Assistant Professor of Management at SKK Graduate School of Business.