The glass ceiling and the variance of narcissism – UPDATE

This piece suggests that the UK may implement quotas to increase the representation of women on FTSE companies. I appreciate the sentiment. Even though it’s hard to find someone who will explicitly state that women are unsuited to positions of power, the corridors of power both in government and in the corporate world are still very much a sausagefest. For those who think that there is nothing in maleness that makes a better leader, the urge to force something through is strong since apparent watersheds in the past have done little – 30 years after Thatcher was elected where are the women in UK politics for example?. Childbirth and associated gender roles may play a part, but only so much.

There’s equity reasons to desire greater representation of women of course – it’s unfair that women may be barred from these positions based on characteristics that appear to have nothing to do with the qualities we’d want in these positions, but I have a more selfish motive. Basically, if we’re excluding a large proportion of the pool of potential leaders based on arbitrary ground, we won’t get the best leaders, and the society I live in will be poorer both materially and otherwise.

This brings me to a speculation about why the glass ceiling is so resilient.  It may be to do (in part) with differing variances in narcissism.

In our society, these positions of power are in hierarchy. They’re not market derived wealth and power but positions within and at the top of pyramids where the numbers thin out as power increases. To be on a board, or to be a CEO, or an MP, you have to select yourself as being, and convince others, that you are one of the very few suited for the job.

The put another way, to want to be a member of the Federal Parliament you need to think you are better suited than the 99.99% of the population who is not in the Federal Parliament. To be a CEO you need to genuinely believe that your judgment is better than similar proportions of the population and that you output is really worth many many multiples of the average worker. If you don’t believe it yourself, how can you convince others, claim credit and self promote?

Even if you can objectively claim greater than average intelligence, judgment etc., claims of this magnitude cannot be considered rational. Only a narcissist could believe them.

Imagine narcissism as a spectrum from the modest to the highly narcissistic. Then imagine that the population is normally distributed along this spectrum. That is, most people are of middling narcissism, but there is a tail of the highly narcissistic, balance by a tail of the highly modest. In fact, you can look at the simple graph at left.

Now imagine that men have a greater variance in their distribution of narcissism – that is to say they have more at either extreme whilst the average man and the average woman are exactly the same (at the green line). This is speculation, but it seems intuitively right. There is greater variance in other aspects of male life compared to women (such as propensity to be rich and be homeless) and in my own experience they are more likely to be at emotional extremes – despite stereotypes of moody women. Still, there’s probably BIG5 research or the like to determine whether this speculation of mine has any basis.

Now say that there is a certain threshold of narcissism that is required to promote yourself into the leadership races. If you’re less self aggrandising than that, you either won’t choose yourself or be able to convince others of your virtue. I’ve marked this on the graph with a black bar. Since is thicker, the number of men in the pool (shaded pink) is far greater than the number of women (shaded light blue). So even if selections from this point on are gender blind (which is probably still unlikely), you’ll still end up with the leadership sausagefest. [fn1]

But narcissism isn’t really correlated to the virtues we want in leaders. In so far that it is desirable in a leader, it is only because our methods of choosing leaders requires it. It many cases it may well be highly damaging, such as when a leader is full of confidence about their own judgment on a risky investment or the existence of Iraqi weapons when objectivity says otherwise. So we’re eliminating a huge number of potential leaders, perhaps not directly on the form of their genitalia, but something just as irrelevant. It is worth mentioning that if this speculation is true, it’s not just the vast majority of women being excluded, but a sizable majority of men. Undoubtedly outright sexism still plays  a role (if rarely expressed), as does childbirth and probably the simple hysteresis of existing leaders promoting people like themselves. But there might be other variances that differ with gender, such as status seeking, which skew the pool of potential leaders with no correlation to ability.

How do we rectify this? I dunno! But I think quotas are probably as good as anything else, or at least no worse than anything else, including the status quo. They can’t be distorting something that is working well (since the evidence for a CEO effect on company performance is terribly weak). I don’t know if there will be a critical mass/watershed effect that would later make  the quotas unnecessary but maybe, just maybe it could end up moving the selection process away from something as arbitrary as narcissism.

UPDATE – Via Don in comments, there’s at least one bit of empirical work that suggests there is a gender difference in mean narcissism, but not in variance as I suggest (at least, the differences in variances appear quite minor, even though men have higher standard deviation on 4 of the 5 traits). This may not change the fact we are selecting based on an arbitrary trait found more frequently in men than women, but it might mean that the trait itself is more typically male, rather than just found in greater numbers.

[fn1] I may as well mention here that is probably takes narcissism to be a blogger and think that other people are eager to hear what you, as just one person, has to say (and withstand the brickbats in comments). This may be why there’s disappointingly few female bloggers even after the lady on the internet factor. Since I don’t think narcissism is associated with insight, just the belief you have insight that others would want to hear, this is a pity.

About Richard Tsukamasa Green

Richard Tsukamasa Green is an economist. Public employment means he can't post on policy much anymore. Also found at @RHTGreen on twitter.
This entry was posted in Economics and public policy, Gender, Uncategorized. Bookmark the permalink.

24 Responses to The glass ceiling and the variance of narcissism – UPDATE

  1. most self-made millionaires are men too. Is that also because of their narcissism? Hard to believe.

  2. Ken Parish says:

    J. Paul Getty once said “The meek shall inherit the earth, but not the mineral rights.” He should have added CEO positions to the qualification.

  3. Andrew Norton says:

    But how do we apply quotas, given that every company has a separate CEO selection process?

    Another way of looking at this is that women have more socially acceptable options than men – few people think less of them if they decide to put family over work – and this inevitably substantially diminishes the number of women who are contenders for the most powerful and demanding jobs.

    In the absence of evidence that significant numbers of willing and able women are being overlooked, the status quo is preferable to yet more regulation.

  4. Andrew Norton says:

    The blogger example should really make you think – even when it could not be easier to participate in public life – you don’t even have to use your real name, you can combine it relatively easily with other activities – the same gendered pattern appears.

  5. D W Griffiths says:

    Hilarious. It was precisely this variant of distributional graph that got Larry Summers thrown out of his job running Harvard.

    Richard, if this is a sort of complex pun, it’s a great one.

  6. Richard Green says:

    Paul – Whilst not downplaying other effects, I still think so. Whilst this departs from the hierarchy effects I focused on, to be a self made millionaire you still need to believe that you will be able to use capital to find an opportunity that other people haven’t to be able to make a return higher than the interest rate – which requires a verge to the narcissistic side – and then you need to believe that enough to convince the bank or investors or other sources of capital that it is true, so the same dynamics are at play. This carries a subsidiary hypothesis that since the narcissism is not correlated with ability, this selection mechanism would mean that the majority of self made bankrupts are men as well. I believe this is true.

    Andrew- The blogger example is one of the things that made me think since there’s always the opportunity to be the Miles Franklin of blogging, it seemed it wasn’t all social expectations – although these are often internalised and why feminisms greatest successes have been achieved when it convinced women rather than men. I figured there must be some other arbitrary selection mechanism.

    I’d generally be more cautious about regulating a solution, but I can’t see much to suggest that the mess of principal agent problems and corporate command structure decision making are providing a something that is any less distorted or more market based than a regulated option. We haven’t even got the remotest piece of evidence to suggest that executive pay is related to productivity, which would be one of the first things to look for to indicate an efficient system.

    DW – I’ll admit that this thought did gestate originally as a facetious response to Summer’s logic. Unlike Summers I don’t see much to indicate that anything that actually relates to virtue has different variances between the sexes, but what started out as a joke to imply that Summers was a narcissist started to make more sense.

  7. Paul Frijters says:


    you essentially presume there is a link between narcissism and the ability to convince others of something. That sounds like a productive trait in many jobs.

  8. Richard Green says:

    It’s only productive in an economy wide sense (as opposed to private gain, where theft is also productive) in so far that you convince them that something is true when it is true – i.e you convince them that a loan is worth making when it actually is, or that you are the best person when you actually are. It doesn’t help when you’re Eddy Groves convincing someone to give you a loan or a CEO taking credit for gains from a resource boom. It doesn’t actually help the economic system at large allocate or exploit resources efficiently. If there was a better selection mechanism to determine who to loan to or chose for a job or better information, the ability to convince would lose the potential for private gain it had without necessarily adding to aggregate production.

  9. Ken Parish says:

    Apropos Andrew’s blogger example, I am aware from speaking with quite a few women who have poked their toes into the blogosphere over the years that quite a few find the conversational style and its (narcissistic?) male dogmatism, posturing and aggression quite alien and unwelcoming. Thus the heavy disproportion of male to female bloggers and commenters might be argued to be a manifestation of Richard’s hypothesis rather than a disproof of it.

    That said, my own biased assessment is that the conversation here at Troppo is invariably extremely civil and welcoming and yet the lack of female participants is every bit as evident as at more combative blogs. When I checked a couple of minutes ago there wasn’t a single female name in the “recent comments” sidebar list. I wonder why?

  10. conrad says:


    I’m under the impression that whilst males have slightly higher narcissim scores than females, the spread in the distribution is pretty similar. Try doing a Google Scholar Search for the Narcissim Personality Inventory.

    “It was precisely this variant of distributional graph that got Larry Summers thrown out of his job running Harvard.”

    I thought Larry Summers was booted because he said it was genetic, not just for saying the distribution was different. I think many people would be happy to have different distributions, since for things like maths, that could be evidence for gender bias in schooling and parenting, not biological differences.

  11. Don Arthur says:

    Richard – Why imagine what the distribution of narcissism looks like?

    Psychologists have developed a number of scales for measuring narcissism — for example, the Narcissitic Personality Inventory (NPI). A quick literature search will give you an idea of what the male and female distributions look like.

    For example: ‘Gender Differences in the Structure of Narcissism: A Multi-Sample Analysis of the Narcissistic Personality Inventory

  12. Richard Green says:

    Cheers Don, I’ve updated the post for the benefit of others, and thanks likewise to Conrad. I had tried a quick search based on the Big5 traits, but I obviously wasn’t looking very well. I’ll look better now

  13. Paul Frijters says:


    for your story to work, not only must it be true that there are more men who are very narcissistic and that narcissims conveys an advantage in convincing people of something, but it must also be true that there is no relation whatsoever in the ability to convince others and the objective value of what a person believes in. That last one is really a very strong and extreme assumption. You need it because otherwise the ability to convince conveys a real advantage on average in the top position. The reason it is an extreme assumption is that in order for it to hold, it must then also mean that others who help you get to the top are completely unaware of the objective value of what you believe in. If those in a position to help others have some small inkling of the true value of what you believe in, then you will find that those who make it to the top are those who can convince others and have been allowed to get to the top because what they happened to believe in was also of some objective value. Hence the assumption of no relation between the ability to convince and having something of worth to sell at the top boils down to an assumption that no-one in an economy is incapable of discerning true value.

    If you allow for some discerning capacity amongst those who co-decide (bank managers, co-directors, co-workers, owners, etc.) then even if there is average relation between narcissism and having good things to sell, there will be such a relation at the top and your story breaks down, i.e. you re then in the business of saying that men rise to the top because their ability to sound more convincing (narcissism) is actually a productive trait when coupled with the hurdles of peer approval that men must clear in order to get to the top.

    My own opinion is that men are more likely to rise to the top because of male bonding: the sisterhood doesnt work as well as the brotherhood.

  14. A female friend was among those urging me to leave Catallaxy years ago because she was interested in my blogging but did not like the uncivil comments threads. So I set up a new blog with a civility policy that is generally observed. But she has only commented two or three times in four years.

    While I have had a few regular female commenters over the years, their overall numbers are very low.

    My attitude to these things is that just about all that can sensibly be done to encourage women has been done, and that ultimately we must respect the decisions they make about their interests and life priorities.

  15. Nicholas Gruen says:

    Thanks for another great post Richard.

    I’ve always thought that the absence of women in politics is in fact a symptom of a larger problem which is the way in which politics is skewed towards a particular personality type – of whatever gender.

    And that’s really a variant of a larger phenomenon which one might call the Groucho phenomenon. (I’m thinking of his comment that he’d never join a club that would have him as a member.)

    There are professions in which the people who want to do them, are disproportionately, by that fact, the wrong kind of people. I expect this is a relatively small sub-set of all professions, but it’s numerous enough. I can think of these.

    * politicians (are they in it to really do a good job or do they just want to be the centre of attention)
    * psychiatrists (are they flakey types who did psych because they wanted to work themselves out and never did?)
    * ‘spiritual counsellors, like priests (are they generally committed to the spiritual or emotional journey or are they dullards who want to do something safe, secure and well thought of by their narrow community.)
    * public servants (ditto, mutatis mutandis)
    * police, jail-warders and security generally (is part of the attraction physically lording it over people?)
    * judges (perhaps) (How pompous are they? How much do they want to see justice done?)

    Note that the desire to do these jobs may contribute to someone doing a very good job, but often the desire is ‘tainted’ with bad qualities.

    Whenever I see people raising the issue of discrimination, I always think of all those kinds of discrimination that we just don’t worry about, just pretty much let go through to the keeper, in favour of height, good looks or even just ‘introversion’. But as you forshadow, it’s not easy to come up with rules which treat these matters.

    Which leads me to say that, while quotas may or not be worthwhile, their obvious problem is that they’ll end up turning up the very most narcissistic women!  I think this is a genuine problem for instance in grooming women for corporate and other kinds of leadership. If one of the benefits one was hoping to get out of it is a different kind of personality type, you may not get very far, and women may begin demonstrating various tendencies one didn’t much like in men.

  16. Maria says:

    Re blogging – maybe women are busy? It would be interesting to see the stats for Facebook to see if there are more women than men, because Facebook is so very easy to use and takes up little valuable clothes folding, school pick up, child play date, dinner making, vacuuming, grocery shopping, dog feeding, lawn mowing, AND a 37 plus hours of paid work time, time. Also, flames. They really put women off. We are, generally speaking, more polite and less tolerant of online rants, don’t you think?

  17. FDB says:

    What the *&#% are you bleating about, Maria?

    Just kidding. Nick’s comment above yours kinda lays out the same argument, only with the depth of detail you’d expect from someone completely unencumbered by household or professional duties.

    Kidding again.

    Or was I?

    As a male, I’ve got plenty of free time to hedge.

  18. Pingback: Club Troppo » Missing Link Friday – 12 November 2010

  19. meika says:

    Randomness is the answer, at least for the Peter Principle, which must be correlated somehow with brotherhoods of narcissists doing better than sisterhoods.

    In the late sixties the Canadian psychologist Laurence J. Peter advanced an apparently paradoxical principle, named since then after him, which can be summarized as follows: {\it ‘Every new member in a hierarchical organization climbs the hierarchy until he/she reaches his/her level of maximum incompetence’}. Despite its apparent unreasonableness, such a principle would realistically act in any organization where the mechanism of promotion rewards the best members and where the mechanism at their new level in the hierarchical structure does not depend on the competence they had at the previous level, usually because the tasks of the levels are very different to each other. Here we show, by means of agent based simulations, that if the latter two features actually hold in a given model of an organization with a hierarchical structure, then not only is the Peter principle unavoidable, but also it yields in turn a significant reduction of the global efficiency of the organization. Within a game theory-like approach, we explore different promotion strategies and we find, counterintuitively, that in order to avoid such an effect the best ways for improving the efficiency of a given organization are either to promote each time an agent at random or to promote randomly the best and the worst members in terms of competence.

    In studying smart swarms it is the little bits of randomness in an ant’s individual behaviour that helps the antnest solve continuously changing problems in its environment. It help provides a diversity of attempts. Bureuacracies, business or government, don’t like that, they like to know what their assigned fools are supposed to be doing all the time. It leads to chaos. Allow a little bit of chaos at the individual level and you get efficient productive order. Bureacracies, business and government, are in love with their vision of Order, they are organizational narcissists, that’s why individual narcissists do so well in them, and psychopaths even better.

  20. Ken, I should probably comment at Troppo more often, but tend to find that I’m kept busy at my own blog. I do think what we’ve achieved is interesting: an all female blog that isn’t feminist (it’s mainly legal, although our interests are expanding now there are four of us) with regular commenters that are not only mixed politically, but evince a balance of genders.

    I don’t know of any other large-ish blog that does this.

  21. Pingback: Club Troppo » Lottery policies – places for transparent arbitrariness

  22. Pingback: Club Troppo » A modest proposal: Affirmative action or reverse discrimination for those who’ve broken their careers to care

  23. Daniel Jeffares says:

    The Glass Ceiling Redefined by Daniel Jeffares

    I disagree with the fundamental premise upon which the debate about gender equality in the boardroom rests. Based on my considerable international experience of working at the director level of major organisations domestically and abroad in more than 20 countries, I can say that elevation to the top eschalons of corporate power depends on a great deal more than one’s competence.

    With few exceptions, the people who attain status within such organisations have had a deep and continuous mentoring relationship with one or more older executives, who sponsor the younger person’s elevation through the ranks as they follow their mentor’s career path upwards. They typically devote the vast majority of their time and energy to immersing themselves in the complex fabric of the organisation’s DNA and serving the interests of their career sponsor.

    Most weeks involve 12 hour days in the office and a handful of evening or weekend commitments to ensure that you only ever see you kids when you drive them to sport during the Saturday morning peak hour where you will spend your day networking with the other parents at your child’s elite private school. Reading to them in bed is a delight to you both but it ends all too soon and you know it could be another three days or more before you see them again though you all inhabit the same house. You’ll be obliged to compromise your ethics when it falls to you to front the media to explain your position on say mining coal seam gas from under our water catchments, representing the interests of your stakeholders ahead of those of your own family and the rest of the community if required.

    After investing 30 years or more of your youth pursuing this path, you might, if you are simply extremely lucky and your network of close contacts built up over literally thousands of work function interactions yields a sponsor who will put you up for the board, you’ve made it.

    I believe it is a career path open to all, should you choose to accept it, and those women currently at the peak of Australian business probably worked harder than most to get there. But let’s not diminish their achievements and debase the status of women in business by nominating a minimum female quota on the public boards.

    I’d suggest that half of the people currently excluded from senior executive appointments by the glass ceiling are men. That’s because, rather than discriminating on the basis of gender, the glass ceiling actually discriminates against those who have not devoted themselves exclusively to their careers. That is an option that is open to us all.

  24. Pingback: Club Troppo » The political economy of unindexed income tax brackets

Leave a Reply to Andrew Norton Cancel reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.