The Old Boys’ Club

It is real, at least in one Asian data set, as these new NBER working paper results are brought to us by Zoë Cullen and Ricardo Perez-Truglia:

We use an event study analysis of manager rotation to estimate the causal effect of managers’ gender on their employees’ career progression. We find that when male employees are assigned to male managers, they are promoted faster in the following years than they would have been if they were assigned to female managers. Female employees, on the contrary, have the same career progression regardless of the manager’s gender. These differences in career progression cannot be explained by differences in effort or output. This male-to-male advantage can explain a third of the gender gap in promotions. Moreover, we provide suggestive evidence that these manager effects are due to socialization between male employees and male managers.

There is more to the abstract, including a discussion of the benefits of smoking together.  Here is an ungated copy.


Could it not be that female managers are not promoting male employees at the rate they should and male managers are better at recognizing talent?
Is that possible?

"These differences in career progression cannot be explained by differences in effort or output."

Maybe if you wish hard enough you can explain it by a gap in talent. Lol

Aslan if you're that willing to take at face value that it's possible to measure effort and output in some generic manner, you're probably going to believe, well, whatever it was the researchers believed before they started on their "research".


I've been a fan on this blog for over 10 years now and know that Tyler has posted a ton of research with conflicting conclusions.

I had the initial first thought as Sk (not that I believed it to be true, but that I wasn't sure the paper ruled it out). But here's what the paper has to say about it:

Another potential mechanism is that male employees work harder and are more productive under male managers than they would be under female managers. For example, male managers might be better than female managers at motivating and monitoring male employees. To test this hypothesis, we exploit data on two measures of effort (the number of days worked and the number of hours spent in the office) and one measure of output (sales revenues). Contrary to this mechanism, we find no significant evidence of a male-to-male
advantage in any of these measures of effort or output.

That sounds like a pretty reasonable explanation to me.

Studies like this are done with a conclusion in support of data. They search for anything that will confirm their bias and discard anything that does not.

I've got a female manager. We bump heads because she's not strong on logic and she "feels that I'm not hearing" that her emotion based arguments.

Maybe, she knows English grammar better ;)

Maybe she knows English grammar better ;)

Have you considered that your female manager may be a "Feeling" type making decisions and judgements based on her personal values, and you may be a "Thinking" type making judgements using impersonal and objective logic? Both are valid types, they just make decisions in different ways.

And to think that describes several male managers I have worked under . they were not strong on logic and felt that I'm not hearing their emotion based arguments.

Or maybe it is simply that there are different types of people, and thinking that only men understand logic is or don't have emotional attachments to their illogical arguments is the sort of thing that requires a certain perspective.

Go team.

It is also the case that logic matters less as you progress up the hierarchy. Or perhaps I should say emotional competence matters more. Something about interpersonal relations, networking, etc. It is really amazing how many leave college thinking that the analytical skills they learned are sufficient to succeed.

Here's what I'd want to know: were female managers slower to jettison underperforming employees, also? In my experience that's kinda been the case, anecdotally. This might just be the case of male managers being a bit more ruthless in their personnel decisions than female managers, particularly when you consider that most of the really high performing employees are likely to be men in the first place, for the same reason there's more men in prison and in homeless shelters: there tends to be greater variance among groups of men than among women along a number of dimensions.

From the abstract, it seems that ambitious female employees should "schmooze" more with male managers; or, in extremis, take up smoking. Sounds about right?

Maybe take up golf, single malt Scotch, skeet shooting, poker, fine cigars.

If that doesn't do it, establish quotas.

It isnt 1980 anymore, hun.

Female employees should join the same dining clubs as their male managers and accompany them on their grouse hunting trips.

Oh wait, it's not 1890 either.

I went grouse hunting in 2018.

Several points stand out as issues quickly browsing the paper:

1) I would question their measurement of "Effort or output" and how generalizable their sample is.

Measurement variables seem like "effort (the number of days worked and the number of hours spent in the office) and one measure of output (sales revenues)"

They note: Contrary to this mechanism, we find no significant evidence of a male-to-male advantage in any of these measures of effort or output.. But does this sample generalize to work as a whole? I thought men often worked longer hours than women, due to other commitments.

Part of the answer may be that for effort they specifically exclude out of work commitments: We subtract the number of approved leave days (i.e. parental leave, sick days, vacation days, etc) from the total number of work days in the month.

They also construct a de-novo customs sales index using a subsample to measure output: Finally, our measure of output is based on the 38% of employees who have a sales role. We measure sales performance based on their sales revenues. The bank uses an official formula to aggregate an employee’s sales over all products (e.g., credit cards, loans, mortgages). We use this data to construct a sales performance index on a monthly basis

What's the evidence that this correlates with how performance is internally measured and how employees orient towards that? Also, how do they know that their subset of employees which doesn't show an output difference is the same subset as is driving their effect?

2) Let's also state that their sample is: "We collaborated with a large commercial bank in Asia that has millions of customers, billions of dollars in assets and in revenues, and thousands of employees.". So at best this sample is, for a headline... "The Old Boys Club is real... in Japan/Korea/PRC/Taiwan/some unspecified East Asian country".

3) Spending more of your breaks with managers sounds like spending extra time effectively on the job, in practice, effectively in meetings. I would not be surprised then, if that results in faster promotion or selects for more ambitious, workaholic employees.

I would bet there are numerous other points to consider.

in Vietnam to be exact.

In section 3.1 they note that they're looking at a single large banking company ... in Asia. OTOH, in that section they show stats for that bank and that country, compared to US banks and workers, and show that the statistics by gender on labor force participation, wages, and higher level management positions are similar for their research subject and the US.


"We collaborated with a large commercial bank in Asia that has millions of customers, billions of dollars in assets and in revenues, and thousands of employees. This organization is typical in that female representation drops off at higher levels of the corporation: 75%of entry-level employees are women, which falls to 61% in middle management, 25% at the C-Suite level, and 0% at the CEO and company board levels."

A quick internet search yields that the average age of a C-suite level executive is 54 years old. It's reasonable to assume these people started their careers 30 years ago. The present conditions are the accumulation of outcomes of 3 decades. So, why try to explain the past with the present socialization/smoking in first place?

Yikes. Yeah, this is significant enough that maybe the original post needs an update? "55 year old managers at one bank in Asia are kind of sexist" isn't really breaking news, and there's not really any reason to expect it to generalize.

After a closer read, I think we could charitably say that the paper is trying to explain some of the mechanical details of how the "Old Boys Club" effect operates: a new male manager takes over a team, some of the male workers begin spending more social time with the male manager over time especially if they both smoke, then after a year or so those male employees experience better promotion prospects relative to their female peers.

Whether this is widespread or a large effect isn't the primary point of the paper. They admit up front that their data all come from a single bank in Asia.

I think Tyler's deadline (and to some extent the title of the paper) set us up for an exploration of the SIZE of the "Old Boys Network" effect, when in reality the paper is more about walking us through the mechanics of how the effect manifests when it is present.

Basically HSBC Or Standard Chartered. Two Brit Banks who make most of their money in HK. Could be guys promoting their school alums over the local gals. How universalizable are conclusions with a sample size of one company?

The most important rule of the The Old Boys’ Club is not to talk about the The Old Boys’ Club - to girls.

Interesting. Do male managers promote more or less overall in absolute terms?

Also, where I work, all promotion decisions have to get made several rungs up the ladder, and budgeting ensures only a fraction of those put forward get picked. I would be very willing to believe female managers have a harder time advocating for their employees and getting listened to in that environment.

When I was in the corporate world, I noticed 4 examples in one firm of high ranking executives who had proteges who looked quite a bit like younger versions of themselves. I was the protege of the tall Chairman, for example.

You are not the only person to have observed this phenomenon. I used to work with a company in the Pacific Northwest that had a diligent, punctilious and mostly female HR staff, and every new executive hire just happened to look like a Canadian mountie.

Nice people, well-run place but not much self awareness. Seemed to work for them.

There are as many explanation for this as there are female managers, and folks will choose the one that appeals to their prejudices (about men and women). One explanation is that female managers are more demanding and, thus, less likely to promote their employees. Without mentioning any names, anecdotes about female bosses seem to bear this out. Another less satisfying explanation is that female managers are insecure and, thus less likely to promote employees for fear the promoted employees will surpass the female manager. Both of these explanations attribute agency to female bosses, which is more satisfying than (a) female bosses are incompetent or (b) the female boss's bosses are sexists. Anyway, as I indicated, there are likely as many explanations as there are female bosses. My only experience with a female boss was my former wife, and she was really demanding and never quite satisfied with my, ahem, performance and unlikely to give me a promotion.

So... female managers discriminate against promoting men? OK, got it.

Not sure why you would spread this information. It's your career funeral...

This objection is addressed upthread. The paper has problems, but that's not one of them.

"Contrary to this mechanism, we find no significant evidence of a male-to-male advantage in any of these measures of effort or output. "

That part upthread? Respectfully, how does that address the idea that this data could be explained by female managers discriminating against males? From the blurb: "We find that when male employees are assigned to male managers, they are promoted faster in the following years than they would have been if they were assigned to female managers. Female employees, on the contrary, have the same career progression regardless of the manager’s gender." That statement is consistent with male managers not discriminating, but female managers discriminating against males. Not saying that is the only explanation, but it does seem like a consistent and possible explanation. The statement in the paper about effort doesn't seem to discount it. Or maybe I am missing something?

I think the problem is

Mike Pence disease.

In law at least, you train associates to become lawyers and more specialized. This involves office and non-office interaction, and you work nights and weekends on project, and travel with an associate. I think there is a handicap in being a female associate. It's easy to assign work in the office and oversee it one to one, but if there is an out of town project you might be inclined to have several people go with you on the trip, or pick a male associate.

The problem may not be as acute when there are more female partners. And, in fact, I have seen the gender divide go the other way: female partners form alliances and market to female in-house attorneys, who have a better career path progression than outside counsel.

It's not easy, and I can understand where female associates feel left out.

Doesn't this imply that male managers promote people more on net than female managers? Like if a female managers is replaced by a male one, males are more likely to be promoted and females see no effect, therefore...?

I can't see how that's possible. Managers are only empowered to promote people when there is a position to promote them into. They aren't in control of the number of promotions available.

In my world "This person is excellent; we should create a higher position for them" is how career advancement happens. The new job will be nominally open to all applicants but the requirements are tailored so that only the specific person they have in mind fits the bill.

So, maybe female managers don't have the connections to still higher level managers necessary to authorize the creation of a new position for people they think are excellent?

"I can't see how that's possible. Managers are only empowered to promote people when there is a position to promote them into. They aren't in control of the number of promotions available."

That's certainly not true everywhere. There is a large differential in promotion rates between offices at my company. One office routinely promotes people to senior level at the 5-10 year level and the main office promotes around the 8-13 year level. There are some technical arguments for why this happens, but it comes down to the branch office is just more aggressive with promotions.

There's not a given level of slots, it's primarily based upon skill level. And of course senior engineers bill out at a higher rate.

And of course senior engineers bill out at a higher rate.

So you could argue that female managers are being more cost-conscious by not giving out as many promotions.

Where I work, there's "Engineer" and "Senior Engineer" which are pretty subjective, but beyond that, to get to "Principal Engineer" or higher, you really have to have people working under you. You don't get it due to years of experience, you have to be leading a group.

I don't care much about this. But. How do men get promoted faster by men,yet everyone else is not promoted slower? Or is that what is meant by "same career progression"? Something about zero-sum games seems relevant here. I don't know much about Asian cultures, but claiming that an Asian business entity is "similar" to an American one based on a couple of statistics is laughable. Then there's the problem of chicken and egg. How did they show that the men and women hired by the bank were drawn from the same population (in terms of competence, drive, and alternative opportunities). I doubt they did. There are at least a handful of reasons to promote someone, job competence isn't one of them.

Re "Why isn't everyone else slower?", assuming Asian bank demographics, female manager->male manager switching, the only pathway they find generates faster career progression, is probably not frequent enough to have an effect on the whole? Male->male manager is probably most common when switching and most employees did not switch manager.

Seems possible there is some artifact of male managers being more senior on average than female and switching female->male switching tending to select for more headhunted men? Or that female and male managers have different specialisms and switching between one of these is associated with a particular career path associated with faster progression?

I think it's kind of an odd design that they use - surely you'd simply be interested if in male managers *as* a whole matching with male employees led to a faster career progression than females M matched to female E or cross sex M+E matches. Rather than actually "conditional on switching effect", which then leads to a much more limited and less representative sampling pool. If we found the subset of males with a female manager who switched to a male had a faster whole career progression across the whole sample than males who had male managers throughout, then that would be inconsistent with their prediction, for ex.

Although re demographics: "The firm may be unusual for the financial sector in that a majority (65%) of its employees are female."

Humans are naturally predisposed to favoring others who share their demographic characteristics, such as race, gender, or nationality. To avoid discrimination, the best solution is to limit human discretion and make decisions based on objective criteria or blind the decision-maker to the target’s demographic characteristics where possible. If these measures are impossible, affirmative measures to counter bias such as the NFL’s “Rooney rule” requiring interviews of minorities are justified to ensure that positions are not simply passed along old boys’ networks.

Unfortunately, affirmative action is most prevalent where it is least needed and indeed counterproductive (in college admissions, where objective criteria like standardized tests are a major factor and it is possible to blind applicants’ demographics as admissions counselors never meet them) and least prevalent where it is most needed (in workplaces where standards are opaque and dependent on connections).

Best comment today.

When AI starts making hiring decisions on past data aka machine learning, watch bias get passed off as objective criteria ("because it's math!").

"It is real..." Sorry, outside of academia, you actually have zero actual experience or understanding here. Same with the paper's authors.

There is another factor. Some women, having reached a certain level of authority, prefer strongly to be the only alpha female on the team and act to ward off perceived competition.

This is not the rule, but it is an exception that crops up again and again, even if we're not allowed to say so.

And people wonder why it is still called the Old Boy's Club, even if we are not allowed to talk about it.

Go team.

Gender gap or agenda gap? Not even "aspirational egalitarianism" does a thing to make "egalitarianism" possible or real.

When might our elites drop their elitist democratic pretenses and shed the pernicious "stratified egalitarianism" that they dangle over all other heads?

Oddly, using a large firm, who shall remain nameless, (n>3,000) and performance reviews/promotion grooming, I found just the opposite. Male bosses were statistically indifferent while female bosses had a preference for women.

More reason for men to avoid female managers.

Anecdote time: my first boss out of college was female and I owe more to her for my career than anyone else. She gave me very good feedback and advice on how to better promote myself to get ahead. 10 years later I'm making 200% more than what I did at that job in a much more senior role. But she never actually promoted me during her tenure (2 years) so she wouldn't show up in this data.

Meanwhile my male boss at my last job promoted me up into the role he left vacant after his promotion - solely because he liked me - and I sucked at that job (managing other people) and had to bail from that company before I was forced out.

People find it easiest to relate to, communicate with, and appreciate the skills of those who are most like them, communicate like them, and have the same skill set. Men find it easiest to work with and for other men. Women and minorities can have different skill sets which are harder to appreciate, but beyond that, women and minorities will socialize and communicate in different ways than men do. There are many barriers to good communication across sex, racial, and cultural boundaries. What we need to be careful of is to separate our appreciation of skill sets from our ease and difficulties in communicating with each other. If you are a man and you find it easier to work with other men, is it because women are lacking in ability, or is it because of communications barriers, most of which may well be your problem (the women probably have no choice but to work with men).

The ability to work with a diverse collection of people with differing skills sets is a skill in itself, one that is often undervalued. The days when "the office" consisted of a bunch of near-identical white guys who watched the same TV, followed the same sports, and laughed at the same jokes is long over. I don't ascribe to the fantasy that diversity is in itself a strength; diversity can be a pain in the ass which creates many barriers to communication and cooperation. But diversity is a fact of life, and those who are capable of rolling with the punches as they work with a variety of different people in the workplace are those who will be the most successful in the long run. White men who only work well with white men are of limited and diminishing value, no matter the value of their other inherent skills. Because of the inevitable diversity in any modern work force, the ability to deal with a diverse workforce is close to the top of the list in the skills of a modern manager. "I don't work well with women (or minorities)" is not only politically incorrect, it is a serious practical shortcoming in any workplace today.


And I think that there are some guys who kind of sense this and that is at the root of some of discomfort with women and minorities and immigrants. They're used to working in workplaces that were near-identical white guys and feel a sense of entitlement to continue having the same status in the workplace without having to accommodate (in the sense of working well and getting along with) women and minorities . I.e. "Why do I have to be all PC? Why can't I just tell the same jokes I did 20 years ago, it's not fair!"

Straw man detected here.

"White men who only work well with white men are of limited and diminishing value, no matter the value of their other inherent skills."

You didn't actually bother reading even the first line of the abstract above, eh?

"at least in one Asian data set"

Comments for this post are closed