Well, we promised more controversy in our paper reviewing, so here we go!
Why are angel investors so biased against women? If you read almost any headline (like this) or podcast setup (like this example) you’ll likely know that women receive much less venture capital investment than their share of the population. “2%” is the number typically used to show men are congenitally biased against women. Why is it?
Today’s paper is one of several that provide some small push-back against these oft-repeated statistics. Women founders get more early-stage capital than these numbers suggest, and in general in more equitable ways. The 2022 Halo Report has female CEOs at 17% of investments (at comparable median seed stage valuations); the 2022 Angel Funders report has similar data, with “female-led companies” at 24-31% of the deals (2019-2021) (and female CEOs raising essentially the same sized rounds as male CEOs, at $296K vs $294K).
That is not to suggest equality of opportunity or total parity in all dimensions yet. But it’s important to be careful with data, so that we are not trying to solve problems that are already being solved while missing real and enduring problems that could have entirely different solutions or implications. It’s likely that the profound division in later-stage VC funding has different causes and mitigations than early-stage angel funding.
We also need to consider carefully where biases might actually be. Which brings that tangent back to what today’s SSRN paper tells us about gender bias in angel investing. First up is Gender, Race, and Entrepreneurship: A Randomized Field Experiment on Venture Capitalists and Angels, from Will Gornall (University of British Columbia) and Ilya A. Strebulaev (Stanford). Available here.
The authors conducted an “audit study,” where participants are real but don’t know they are in an experiment, which is a particularly powerful experimental setup. They are widely used in other fields (like recruitment) but have not been common in studies of angel investing. The authors painstakingly sent 80,000 fake cold email pitches to 28,000 angel and VC recipients (someone has to do it). They randomly selected a male or female name, and a White or Asian name, for the fictitious entrepreneur sending the email. What do you think happened?
If you guessed that males got a higher “interest reply” rate, you would be wrong. In fact, female entrepreneurs had a 4.47% positive response rate; males, a 4.08% rate – roughly 10% less.
If you guessed Asian-sounding names got a lower response rate, you would also be wrong; Asian-sounding entrepreneurs got around 6% more positive response rate.
The results were similar for every subset of investor (male or female), investor type (VC, experienced angel, novice angel), and pitch type (B2B, B2C, healthcare, and whether the pitch was “feminine pitch” or “masculine pitch”), though the statistical significance was different for some (better for angels, B2B) because of the number of respondents.
Our gender and race results are surprising in light of existing studies in other fields (p.7) say the authors. The results surprised the authors, us, and, I would guess, you too.
The paper examines possible reasons, including romantic interest from male investors and homophily (only funding people of the same gender/race) but concludes neither applies. It also (section 5) discusses possible theories of (positive) discrimination, and related explanations. But overall, the conclusion is that early stage investors are not biased at this point in the fundraising process.
What are the implications? Of course this doesn’t eliminate the possibility of bias overall. Bias might instead be from how venture capitalists and angel investors source deals (often from networks and referrals). And/or bias may come during the later stages of pitching and diligence. In fact, bias from those sources might be worse than we feared, because “all” the bias must be here if it’s not in the initial screening stage.
Next time we’ll look at a paper that should concludes male investors are biased against female founders after all – so tune in then for the rest of the discussion – but for this paper we’ll leave you with the surprising conclusion that In that context, our most important result is ruling out discrimination against females by VCs and angels at the initial contact stage of the investment process, as can be discerned from their reply rates to unsolicited email pitches. (p.46).
In addition, a couple of unrelated fun nuggets stood out to me:
- Cold emails work at around 4.5%. Is that more or less than you’d think? Perhaps, in the authors’ words, our pitches are better written and structured than most email pitches received by the typical early-stage investor (p.26). Based on the cold emails in my Trash, this seems likely! Don’t neglect cold emails, and make them good.
- Positive replies come faster than you think: 72% of positive replies are made within a day, and 25% within an hour (p.36)! If you’re going to get a positive reply, it’ll come quickly.
- Asking for a pitch deck does not seem like a natural way to initiate sexual or romantic contact. (p.44). Also accurate!
(If this experimental technique reminds you of experiments with discriminatory reactions to resumes, it’s because of the similar method. One might say another way to compare those results is to say angel investors are less discriminatory than HR managers!)