Study finds women directors damage profits | Hacker News Reader

Study finds women directors damage profits

ft.com

18 pointslimist36 comments

I got a hold of the paper itself, Women in the boardroom and their impact on governance and performance by Adams and Ferreira, Journal of Financial Economics (currently an accepted manuscript, not yet published). It's long, at 46 pages including data tables. Here's the abstract:

"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."

The pre-print PDF seems to have encryption/defense against copy/paste, so I can't post further extracts here.

The pre-print PDF seems to have encryption/defense against copy/paste, so I can't post further extracts here.

Use a better, non-restricted PDF reader? :S

FoxIt Reader beats the crap out of Acrobat: http://www.foxitsoftware.com/pdf/reader/

Suggestions please? I tried Foxit Reader for Linux, no dice, c/p still picks up gibberish. pdf2txt can't extract it. Printing to a ps doesn't help. Would be very curious to hear of hacks around this restriction.

One possible explanation: Companies which are under pressure to diversify their boards select people who are less competant/experienced.

Assuming that's true, why is there a shortage of skilled women directors?

This result would be unsurprising in, say, the 1970's, but in 2009, it is somewhat mystifying, though perhaps the median age of corporate directors is still high enough that an older generation is the norm, before many women enrolled in MBA programs or had lengthy corporate careers.

I think you nailed it there: the median age is still high.

Also consider that the gender gap will likely not disappear for a long time, if ever. Many extremely smart, well-educated women leave the workforce for a period of time because they place the opportunity of raising one or more children above the opportunity to use their skill to continue the corporate war to the top. All else being equal, someone who leaves for a number of years before returning will have a tougher time making it to the top than someone who does not leave.

This could partly explain why although there are more women graduating with undergraduate degrees than men, roughly one third of MBA students are women.

It will be interesting to see what ratio the senior execs drift toward as institutional sexism begins fading with older generations. Probably not 50:50, but maybe closer to what's seen at MBA schools.

A compensating factor could be the increasing percentage of women in college.

http://www.prb.org/Articles/2007/CrossoverinFemaleMaleColleg...

The article itself contradicts that explanation: "women were more effective at tasks such as executive supervision and monitoring", "Female directors also had a better record of attending board meetings", "Our research shows that women directors are doing their jobs very well".

In fact, they also say that female directors were a benefit to "poorly-managed" companies. So, one thing you might claim is that this fits in with the general theory that the performance of men has a higher standard deviation than the performance of women.

Perhaps better "executive supervision and monitoring" is not desirable.

Not if you're the executive in question. Shareholders might feel differently about this :)

Perhaps more closely 'executive supervision and monitoring is undesirable when the monitoring is done by people bound from the top, such as shareholders on boards, or top executives on underlings.'

From the abstract: "This negative effect is driven by companies with greater shareholder rights. In firms with weaker shareholder rights, gender diversity has positive effects."

Yet while those traits often helped badly governed companies, the Ferreira and Adams study suggests that increased monitoring can have a negative effect on well-governed businesses.

That seems to be the key sentence. Although I am not certain, it seems likely that it is board interference with already effective management, not the presence of woman, that forms the causal link here.

Assuming that this study is well run and this popular article isn't oversimplifying... which are very big ifs... some other ideas:

- Companies in trouble might turn to women as board members. Women executives are still mostly outsiders. They might be more likely to join (or be solicited to join) the board of a company doing poorly.

- It's pretty well established that men are more comfortable with high risk -- sometimes to their detriment. Perhaps those companies with more female board members are less profitable but they go bankrupt less often.

- It might be that companies start thinking about board diversity once they've plateaued.

This study by Catalyst may contradict your first proposal -- it shows quite a clear trend between increasing Fortune Rank and an increasing percentage of women directors (and corporate officers, to a lesser extent.)

http://www.catalyst.org/etc/Census_app/07COTEUS/C_appendix5....

They probably act more ethically and as a result sacrifice some opportunities for increased profit.

> They probably act more ethically

You don't have much experience with women.

Another possibility: companies with more women on the board manage their risk more effectively, sacrificing some short-term profit for long-term stability.

Really, I couldn't decide from the article, but we've seen q good number of big companies recently that looked quite profitable until they went spectacularly bust.

Any manipulation with have this effect. If the government said, "You must hire people who prefer strawberry ice cream over chocolate ice cream", even though there is likely near-absolute equality between those two groups, companies will still be damaged by a reduction in their ability to hire freely.

And it will be even worse if there isn't any near-absolute equality between two distinct groups, which tends to be the case in most real-world government preference scenarios.

I guess I implied that that was true, but also not necessary.

The title of this submission is grossly inaccurate---the study in question shows nothing of the kind.

This is a textbook example of the media (and the submitter) jumping on correlation and confusing it with causation. Women represent a tiny minority of board members overall, and it's reasonable to assume that the old boys' network would be strongest among the wealthiest and most powerful boards. This demonstrates nothing but that the glass ceiling cracks on the lowest levels first.

Here is another explanation: Underperforming companies are more desperate and so choose candidates with good track records even if they are not male. Males who can have the pick of the crop are more likely to work for successful companies. So the causation could go the other way: female directors are more likely to be chosen by underperforming companies than overperforming ones.

Your argument requires a preference among successful companies for male directors, which is essentially the same as the "old boy's network" theory.

It may be that the preference is rational, but this study provides no evidence to prove that assertion.

No, incorrect. There is weak correlation between posting an article and personally agreeing with its title, because many times titles are simply re-used, for expediency and accuracy.

Also, I personally did not confuse correlation with causation, see my earlier comment. Sufficient correlation suggests further possibilities and questions of causation. The researchers' quotations strongly suggests they too did not confuse correlation with causation.

The title of the article clearly and unequivocally states causation---it says quite plainly that women directors are the cause of the lower profits, when in fact the exact opposite may be true. I don't doubt that the researchers are more responsible than those deliberately distorting their findings in order to draw an audience.

Regardless of who is responsible, the title remains factually, objectively incorrect.

Another alternative explanation: Female directors disrupt the existing, and lazier, board atmosphere with their better attendance records and stricter monitoring. Social friction, or even a low level power struggle emerges, in the worst cases, directing attention from other directors toward the immediacy of the contentious social situation on the board, in the worst cases fueled by sexism, but possibly just fueled by cultural differences.

(An older version of) the actual study is here:

http://hermes-ir.lib.hit-u.ac.jp/rs/bitstream/10086/15730/1/...

The abstract is:

Although some argue that tokenism drives the selection of female directors, we show that they have a significant impact on measures of board effectiveness. In a large panel of data on publicly-traded firms from 1996-2003, we ﬁnd that (1) the likelihood that a female director has attendance problems is 0.29 lower than for a male director, (2) male directors have fewer attendance problems the greater the fraction of female directors on the board, (3) ﬁrms with more diverse boards provide their directors with more pay-performance incentives, and (4) firms with more diverse boards have more board meetings.

We also show that the positive relationship between corporate performance measures and gender diversity documented by previous studies is not robust to attempts to address the endogeneity of diversity. Instead, the average effect of gender diversity on both market valuation and operating performance appears to be negative. This negative effect is driven by companies with greater shareholder rights. In firms with weaker shareholder rights, gender diversity has positive effects. Our results suggest that diverse boards are tougher monitors. Nevertheless, mandating gender quotas in the boardroom may not increase board effectiveness on average, but may reduce it for well-governed firms where additional monitoring is counterproductive.

The fact that diverse boards perform better than homogeneous boards where shareholder rights are weaker seems consistent with the hypothesis that the diminished performance of women directors is stress and contention based: perhaps the tensions between the male and female members of the board, the CEO, and the shareholders, are more difficult to work within than if the board needn't bow to the shareholders as much.

“Our research shows that women directors are doing their jobs very well. But a tough board, with more monitoring, may not always be a good thing.”

Hm. Let's think about this. Perhaps if AIG had more women on the board, they would have been more closely "monitoring" all the exposure the company had in its insurance of mortgage-backed securities. So maybe AIG would not have made so much money in that line of business prior to the collapse of those assets, but perhaps they also would not have needed a huge government bail-out either. Just speculating here...

The study claims that women on boards tend to monitor more closely and perhaps that's because they are more risk-averse. I can't see that as always being a bad thing. Could be taken too far, but surely it's been lacking in recent times.

That's some big speculation about AIG, and on the part of having women on a board. What makes you think the board didn't know what was going on? They put, and kept in, the CEO, one of the biggest reasons for AIG's collapse. And besides, it wasn't just their over exposure to mortgage backed securities. It was endemic of the entire company.

Here is a great piece on AIG's collapse, it was previously on HN, and is the best article about it I've read yet: http://www.vanityfair.com/politics/features/2009/08/aig20090...

A good link about study design, worth applying to almost any news story about a supposedly scientific study.

http://norvig.com/experiment-design.html

Most "studies" don't even have an experimental design, so they need work right from the beginning.

Anyone find any holes in the data or methodology?

How could we, since the article doesn't provide any meaningful information about either?

Of course, one can point out that they aren't doing randomized trials, but that seems a touch unreasonable to demand.

The article links to the study:

http://www.catalyst.org/publication/13/2007-catalyst-census-...

I didn't have time to read the actual paper. I'm guessing the paper ignores any other context aside from hard numbers within a specific time period; so long term plays (with short term losses) are ignored.

No, that's a previous study which reached different conclusions from the study which is the primary focus of the article.

Stray thought: If "forced" to diversify, then they are probably choosing people they don't even like (much less trust) and this will hurt performance no matter the gender, age, skin color...etc...of the person they feel was foisted upon them.

It is because they're forced to wear pants.

edit: It is because they're forced to wear pants by a chauvinistic structure.

Reply on news.ycombinator.com