Women on boards: strong narrative, weak evidence

Via William of Ockham, this article:

Bitcoin legal expert Carol Van Cleef recently told the World Economic Forum that at the North American Bitcoin Conference she noticed only five of the 300 to 400 delegates were women.

Here’s my theory: women tend to prefer, and do better in, large firms in established industries rather than in small organisations in emerging sectors.

For a woman attending a corporate event, there is nothing more glaringly obvious than being in the clear minority. There are times I have been at events when half the women present are wait-staff rather than guests.

Similarly, I’m usually the tallest person in every corporate event I go to. Where are all the other giraffes?

It’s a familiar story for many women who work in male-dominated industries. Some of the top tech companies in the world, like Facebook, Microsoft and Google, have less than 20% of tech roles filled by women and only about a quarter of their leadership roles.

Yeah, James Damore provided an explanation for this and it cost him his livelihood.

The reality is out of step with the data. Research shows that diverse teams bring higher returns, better long-term strategy and increased workplace engagement.

Firstly, I think the sources in that link deserve a thorough going over, but not in this post. Secondly, are we supposed to believe these giant tech companies, who are happy to sell people’s personal data to the highest bidder in order to steal an inch on their rivals, are leaving millions on the table because they don’t want to employ more women? This seems unlikely.

For tech companies that are striving to become the next unicorn, or at least hold out until the next funding round, it’s a wonder they aren’t clamouring for more women, more diversity and more innovative thinking.

Why yes, whatever can explain such strange behaviour?

Jeffery Hollender, professor of sustainability at NYU Stern, told Business Insider:

“The truth is you will perform better financially by doing things like having a great sustainability program, by having women on your board and in your senior management and by treating your employees well and ensuring that they’re owners of the company. Those things translate into better financial performance.”

If you follow the link, the basis of Hollender’s remarks is a study from 1986(!) which compared the performance of 45 companies that encourage employee stock ownership with similar companies that don’t encourage employee ownership. They found stock ownership schemes had “positive effects” on measures like growth and worker participation. So clearly we need to employ more women.

While it’s easy to understand the importance of diversifying investment portfolios some struggle to grasp the same need to reduce risk by diversifying your board membership and leadership team.

While it’s easy to understand you shouldn’t carry all your eggs in one basket, some struggle to grasp the need to carry a cow pat along with the eggs.

A recent study showed that while “homogeneous groups felt more confident about their decisions than diverse groups, the former groups’ decisions were more often wrong compared to those of diverse groups.”

If you follow the link you find that claim is based on this study, whose executive summary says (emphasis mine):

MSCI ESG Research’s research shows that companies in the MSCI World Index with strong female leadership generated a Return on Equity of 10.1% per year versus 7.4% for those without (as of September 9, 2015, measured on an equal-weighted basis), confirming previous research, though we could not establish causality.

Note that the MSCI World Index consists of giant corporations in each country; the article began by complaining of the dearth of women at a Bitcoin conference. Rather than the MSCI’s study being evidence that having women on boards brings about superior performance, I contend it shows simply that large, established corporations with plenty of money now employ more women. The other study cited in the linked article says:

Women’s presence in corporate leadership is positively correlated with firm characteristics such as size as well as national characteristics such as girls’ math scores, the absence of discriminatory attitudes toward female executives, and the availability of paternal leave. The results find no impact of board gender quotas on firm performance, but they suggest that the payoffs of policies that facilitate women rising through the corporate ranks more broadly could be significant.

Which is hardly going to send executives rushing to hire more women, is it? Back to the original article:

“We decided to put the work into surfacing amazing female candidates. We don’t believe in saying, ‘Hey, I’ll hire the best candidate from whomever shows up at my front door’,” Weebly cofounder and CEO David Rusenko told Business Insider. “It pays to put in a little more effort.”

Surfacing amazing female candidates? Isn’t that the sort of thing that landed Harvey Weinstein in trouble?

Deloitte matches senior leaders with ambitious female talent to not only get exposure to the work of the senior team but to create visibility across the board for the mentee’s skills and future trajectory.

So bright young women get showcased around the company by men twice their age? How progressive.

It also encourages women to expand their networks by leveraging those of the senior team throughout the program.

So how do young women go about applying leverage with older men?

And for many people, having the ability to tap into new networks in an encouraging environment can put women on the fast-track to a leadership role.

As Kamala Harris found out.

Sheryl Sandberg, who famously told women to “Lean In” to their careers, told USA Today:

“Endless data show that diverse teams make better decisions. We are building products that people with very diverse backgrounds use, and I think we all want our company makeup to reflect the makeup of the people who use our products.

I’m glad she’s not CEO of a company making anti-malaria drugs, or Boeing for that matter. And let’s remind ourselves, Sandberg didn’t get to run Facebook by being any good at tech. Yet her mere opinions are cited as evidence corporations need more diversity to succeed. This is what passes for business analysis in 2018.


31 thoughts on “Women on boards: strong narrative, weak evidence

  1. Sad but true. Women are judged on looks, and men on competence. Pretty women get more interviews and promotions than plain janes EVEN WHEN THE HR IS DONE BY WOMEN.

    This over-promotion holds true from bog cleaning to boardroom.

    And if women are selected for looks, it’s mathematically certain that the proportion of incompetent women will be greater than men.

    When did you last see a gushing profile of a corporate amazon, and she wasn’t easy on the eye?

  2. Currently fashionable nostrums get pumped by confirmatory data…………………’twas ever thus.

    In the Eighties the Bank of England was very keen to push the idea that a plentiful supply of non-executives on quoted boards was a good and necessary thing. For their sins, I was hired to put together the report and – wonders to behold – I produced a statistically dense report which ‘proved’ exactly that.

    If the Bank had, instead, wanted to demonise non-execs and show that they were a bad and unnecessary thing, I would have had no trouble producing a report – with all the pretty pie charts and graphs – ‘proving’ that.

  3. Two questions worth exploring (genuinely asking too, it’s not me being mischievous);

    1. If women are paid less than men for the same job, and are equally competent, why aren’t corporations hiring 100% women?
    2. Did companies with diverse boards and management become successful before or after the diversity?

  4. less than 20% of tech roles filled by women and only about a quarter of their leadership roles

    Sounds like women are doing very well. They’re skipping the boring rank-and-file work and heading straight into management.

  5. Competent men actively and preferentially seeking not-so-competent women to run a business always reminds me of a cartoon I saw when a mere toddler. No idea which animation producer did it, but there was a skull of a long dead Longhorn by a poisoned waterhole and its bony jaws opened to sing out: “You’ll be sorrrrrry!”

  6. Recusant,

    In my line of work, we call that “decision-based evidence making”, and keep a wary eye out for its practicioners: there are those who will happily ask what the desired decision is, and produce a study that “proves” it’s the only credible course of action.

    As one of the few good lines from “Robocop 2” put it, “Sir, whether the evidence exists or not, I’m sure I’ll be able to find it…”

    It’s usually possible to spot and dismember such work, given time – but some fairly major and expensive failures like invading Iraq and getting mired in Afghanistan, Nimrod MRA.4 and the 2010 carrier-redesign hokey-cokey have happened when that scrutiny has been skipped. (Which oddly, never seems to have consequences for the protagonists…)

  7. William of ockham, great questions…I’m not so arrogant that I think I can answer them but can add:
    1) during aparthied, white business owners were constantly looking for ways to hire more black people despite laws and quotas to prevent this because they were cheaper. Yet the gender pay gap argument implies that employers are so prejudiced they refuse to have a primaily female workforce, even though they could save 30% on wage costs… Do we really think that discrimination against women now is stronger than discrimination against black people during aparthied?
    2) this would be insanely easy to test. Compare individual company performance over time (whatever metric you want) and compare the gender diversity on boards. See whether diversity preceded high performance. This would be easy to do, all companies would have the data available for this sorr of analysis. But I don’t think it’s being done…

  8. Jason

    Agree with all that. My weak and pathetic excuse is that I was young and needed the money.

  9. “We show that female directors have a significant impact on board inputs and firm outcomes. In a sample of US firms, we find that female directors have better attendance records than male directors, male directors have fewer attendance problems the more gender-diverse the board is, and women are more likely to join monitoring committees. These results suggest that gender-diverse boards allocate more effort to monitoring. Accordingly, we find that chief executive officer turnover is more sensitive to stock performance and directors receive more equity-based compensation in firms with more gender-diverse boards. However, the average effect of gender diversity on firm performance is negative. This negative effect is driven by companies with fewer takeover defenses. Our results suggest that mandating gender quotas for directors can reduce firm value for well-governed firms.”

    Far better paper than the crud you found.


  10. No one really has a good handle on the causation stuff unfortunately. What one really needs is a good instrument. This paper shows a great study of whether family succession or external appointment is better for family businesses. It relies on the fact that family succession is far more likely if the first born child is male. So at least 20 years before the decision is made, a random event (first born child) allows us to infer causation. And family succession is bad.


    Earlier studies were mixed – some found performance good, some bad. But since succession decisions are endogenous (firms doing well are far more likely to have family succession), causation was difficult.

  11. @recusant. We should have a real world chat.

    The Bank of England’s latest missive is that diversity of NEDs improves corporate governance and business performance. Whilst I am sure no-one there would want to go the full James Dalmore I would bet that there are some dissenting people internally on that and on that other charged issue, climate change.

    It’s always the true believers who push this stuff with, or without, evidence. Careers are broken standing up to it and made clinging on. What would you do?

    Never underestimate the difficulty in getting someone to understand something when their livelihood depends on them not understanding it. The same applies here. Facts are irrelevant

  12. We are such a vastly wealthy society that we can afford lots of stupidity.

    When the times of troubles come again, probably from our over indulgence in fun stupidity, all of this weirdness will go away.

  13. Men and women: stop providing “help”, “encouragement”, “visibility” etc based on what sort of stinky bits reside between someone’s legs. I’m utterly convinced this madness will not abate until men hold women accountable in the same manner they hold other men. Men are expected to perform, full stop. Everyone knows exceptions to meritocratic evaluation (old college friends, nepotism and the like) is a fool’s errand, and men must have enough backbone to shun the diversity exception as well.

    I do not have much hope that this will happen, given man’s innate flaws concerning women. Oh well.

  14. “The reality is out of step with the data. Research shows that diverse teams bring higher returns, better long-term strategy and increased workplace engagement.”

    not the first time I’ve said it and it won’t be the last either, but this is another one of those feminist doublethinks. Why are women “diverse”, unless they think differently than men do? Hey, isn’t that sexist? More importantly, maybe the way men think is better for business? Oh, that’s definitely sexist.

    Notice also the people not considered diverse…Apple’s black head of diversity noted that a room full of white men could be very diverse, and then she got fired.

  15. 1. If women are paid less than men for the same job, and are equally competent, why aren’t corporations hiring 100% women?
    2. Did companies with diverse boards and management become successful before or after the diversity?

    1. Lots and lots of executives at companies hire based on things like whether the hiree went to the same university as them, plays the same sports, is related to the boss, etc., etc. The reason this isn’t a good retort is that lots and lots of companies, large and small, do not hire based purely on merit and performance. The other reason this isn’t a good retort is that conscious misogynists don’t pay women 30% less at all levels just because they’re women; they believe women do inferior work. The misogynist believes that’s fine for menial or clerical work, but for anything critical he’s going to hire a man. The end result is a preponderance of women in lower-paid menial and clerical jobs and a preponderance of men in higher-paid professional jobs. Since that’s exactly what we see in the real world, you haven’t disproved anything.

    2. I’m of the same mind as Tim; successful and profitable companies can afford to waste money and effort on public virtue signaling. I think a better metric might be turnarounds of struggling companies, where any misstep or waste of resources is an existential threat.

  16. @Daniel Ream

    One major reason why people hire people like them (eg went to same school/university) is that it is tied to statistical discrimination – you hire based on the information you have. It’s easier to tell who is good if you have the same background.

    “Arrow and Phelps proposed models of discrimination based on rational optimising behaviour and limited information….Typically, in statistical discrimination models, economic actors attempt to assess some characteristic of individuals based on limited information. A classic example is an employer assessing the expected productivity of a worker he is considering hiring. The employer is privy to some imperfect signal of a worker’s productivity, such as the impression he gets in an interview or by reading the job applicant’s resume. He is also aware of the job applicant’s group membership, such as his race, gender or ethnicity. In the absence of perfect information, the employer’s optimal prediction of productivity is a weighted average of the individual‐specific signal he receives and the average productivity of workers in the same group as the applicant..”


  17. “Surfacing amazing female candidates? Isn’t that the sort of thing that landed Harvey Weinstein in trouble?”

    I think that’s more Mitt Romney’s thing – because it sounds like they have ‘binders full of women’.

  18. Since pretty women do get hired and promoted the question should be : Why doesn’t a company that hires uglies not clean up?

    No one does this reverse discrimination. The obvious reason would be that having a munter in the next cubicle is not inspiring, but those neanderthal corporates will outperform for a beauty.

    In other words, more diversity (women) in the workplace could be beneficial, but only as a second order effect.

  19. “…you hire based on the information you have. It’s easier to tell who is good if you have the same background.”

    Women are harder to read and better at dissembling than men. I’ve been on mixed interviewing panels where my female colleagues have vetoed a female candidate (of whom I was slightly doubtful) because they maintained she was lying.

  20. @theophrastus

    Hence statistical discrimination. It is possible that having good women allows you to hire/promote more good women. It’s one of the reasons that I favour promoting more women- they are better at vetting others with their background.

  21. @Ken

    Bullshit. You favor promoting someone based on having a snatch because “it’s possible” that the ‘good’ ones can spot other ‘good’ ones, and that you’re incapable of telling them apart due to your bepenised state. Based on your own reasoning everyone should just hire men since they don’t require an inquisitor to determine their merit.

    Unless you, you know, were kidding.

  22. @zut alors! on October 11, 2018 at 11:37 am

    Sad but true. Women are judged on looks, and men on competence. Pretty women get more interviews and promotions than plain janes…

    Except in politics and public sector – there ugly, fat, robotic, humourless & scowling are positive attributes.

  23. @sam

    No, I favour promoting women because women are more likely to be able to read another woman’s CV and ask the right questions. An ex-public schoolboy who went to Oxbridge can read the CV of other public schoolboys who went to Oxbridge far better than they can read the CV of a state school kid who went to a red brick. Hence statistical discrimination.

    At the initial hiring step when we know nothing, it helps to be able to judge how important being captain of the 2nd XI or being Librarian at the Oxford Union is. Then ask probing questions.

    Men do not share the same life experiences as women, hence less information.

    It’s one of the reasons why anti-discrimination legislation inFrance led to more discrimination by less discriminatory firms. When identifiers of racial minority status (name/address) are removed, minorities were more likely to be rejected at firms that were previously less discriminatory. Why? Because minorities had periods of unemployment that in white candidates indicated imperfection but were normal for minorities. Less discriminatory employers had previously been able to call these people in, now with less information they dropped them.

  24. Daniel Ream,

    “2. I’m of the same mind as Tim; successful and profitable companies can afford to waste money and effort on public virtue signaling. I think a better metric might be turnarounds of struggling companies, where any misstep or waste of resources is an existential threat.”

    Dotcoms with stratospheric growth are almost entirely male. They only get women when all the craziness is over and they’re just a regular corporation.

    These companies hit a certain size and the exciting, independently-minded types quit. These are the people who transform companies. They find innovation and pursue it because the distance from them to the owner is small enough that everyone gets his incentives (basically, making money). When a company gets big, the rules and targets go through so many layers, that gets lost. They become bureaucracies and the innovators don’t fit into that.

    The people who work well with bureaucracies are women. Women will do whatever rubbish you tell them to do without questioning. They rarely bring about improvement like men do. They run a decent ship. The problem is that they don’t improve.

  25. @Bloke on the M4

    Causation here is difficult. It may be that social norms and gender stereotyping has led to this. There is some evidence from some of the psych literature that suggests there are differences in the way that people are wired based on gender, but the extent to which this drives results vs. the stuff that happens in society is difficult to establish – I personally tend more to the view that men are more “techie” and are more likely to be naturally inclined to be innovators, but the evidence is no more than “probably”. I dislike the “diversity is good” slogan because it really doesn’t do justice to the nuances of the debate – see the first paper I linked to above on the positives and negatives of women on boards.

  26. @Ken

    So say you have a few men in place, whom by all appearances achieved their positions via merit. By your reasoning these men cannot properly hire women subordinates, much less identify and groom gems among these hypothetical female subordinates because they are men, and have had different life experiences. The result? No women hired. Either that or you’d have to hire a special Vaginal Consultant to properly vet your hiring – which, come to think of it, is one of those jokes that will probably actually happen.

    Diversity in life experiences can result in poor hiring practice due to wanting something familiar, so why would you exacerbate the problem by further shifting focus from “can he/she get the job done”? In my experience sorting talent from trash is more art than science and – male or female – people either have that ability or don’t.

  27. Given that wives mostly outlive their husbands, all (well, most of) the wealth of Western societies will end up in female control. Why don’t these rich widows help their sisters and simultaneously magnify their wealth?
    Incidentally it is very hard to get facts on this inconvenient wealth distribution.

  28. @sam

    Where candidates stand out, they are hired no matter what the background (at least in most rational firms). But, where candidates look the same people tend to use statistical discrimination. This is a major constraint at the beginning of careers when the information content is low and everyone looks about the same. Also true for career switches and early stage promotions, although there is more info here.

Comments are closed.