A recent study sponsored by TIAA-CREF found that having three or more women on the board enhances corporate governance. The study was based on interviews with 12 CEOs, 50 women directors, and seven corporate secretaries of Fortune 1000 companies. The study found that women impact board governance in at least three ways, (1) by bringing different perspectives into boardroom discussions, including the perspectives of multiple stakeholders, (2) raising difficult issues--that is the study found that difficult problems are less likely to be ignored when women are in the board room, and (3) by altering the dynamics in the board room to create more open and collaborative discussions, thereby allowing management to hear board concerns without feeling defensive.
The study confirms the work of scholars who contend that enhancing boardroom diversity will benefit corporate decision-making. In this regard, the study builds on work by others who claim that boardroom behavior and dynamics can have an impact on corporate decision-making and hence the bottom-line. The study also may be viewed (at least tentatively) as supporting the proposition that enhancing board diversity may prevent corporate misdeeds.
However, the study does warn that the positive impact of board diversity cannot be realized unless there is a critical mass of women directors. Hence, the study found that the number of women on corporate boards make a difference. While a lone woman or even two women may feel excluded or marginalized for their views, the magic number is three or more women directors because at that point having women on the board becomes a normal state of affairs. When that occurs, women directors are not viewed as voicing the "woman's perspective," but rather they are treated as individuals with different styles and perspectives. In this sense, the critical mass of three not only allows women the ability to voice their opinions without fear of backlash, but also ensures that their voices are heard and taken seriously.
I have always been interested in the topic of board diversity, and particularly the impact of that diversity on corporate governance. Indeed, after the corporate governance scandals, there was a small, but vocal group of people claiming that we should focus on board diversity as a way to counteract the "groupthink"/rubber-stamp mentality that seemed to have plagued corporate boards involved in misdeeds. I was skeptical, but largely because of the critical mass point. This study confirms my intuition that critical mass is key. In that sense, it also confirms my intution that we may be exagerating the impact that diversity can have because there are so few companies with that critical mass. Thus, while the vast majority of Fortune 500 companies have a woman on their board, according to the study, only 14.7% of the total board seats are held by women and only 76 Fortune 500 companies have three or more board members.
On the one hand, we should keep in mind that having even one woman directors is important because such directors make significant contributions to the board. However, to the extent we expect women to alter board-room culture, that cannot occur, if at all, until companies actually have more than one.
Thanks to Jeffrey Bauman and Elizabeth Brown for bringing the study to my attention.
TrackBack URL for this entry:
Links to weblogs that reference Is Corporate Governance Enhanced by Women Directors?:
1. Posted by anon on November 9, 2006 @ 11:16 | Permalink
Doug Branson from Pitt has a book forthcoming from NYU Press on this subject: http://www.nyupress.org/books/No_Seat_at_the_Table-products_id-4868.html.
2. Posted by Matt on November 9, 2006 @ 14:10 | Permalink
I'm confused. They proved that women directors benefit companies by ... interviewing women directors? Seems like a biased sample.
3. Posted by Kate Litvak on November 9, 2006 @ 19:57 | Permalink
I actually read the abstract to which Lisa linked, and I see absolutely no evidence to support the article’s bold claims. Apparently, this “study” contains no regressions that link companies’ performance to the appointment of women directors, much less disentangles the endogeneity problem (particularly good companies might be appointing directors as a PR effort, not bad companies appointing women directors and improving as a result). As best I can tell, the paper contains nothing but speculations by a few dozen corporate execs. I can whip our these sorts of “studies” weekly. What a bloody embarrassment.
Incidentally, I’ve seen no evidence of any sort showing that anything about board composition of US companies impacts performance. Lisa, if you’ve seen such evidence, please drop a few links. I will be happy to provide public commentary on those articles too.
4. Posted by Vic on November 10, 2006 @ 0:12 | Permalink
Kate, I think Sanjai Bhagat has a paper finding a correlation between the size of the equity stake of (independent?) directors and firm performance. I can't seem to find it on SSRN, but I'll link to it when it's public. He had a pretty sophisticated was of dealing with endogeneity that was way over my head.
5. Posted by Lisa Fairfax on November 10, 2006 @ 8:38 | Permalink
Vic, I recall that study and feel like I cited it somewhere, but cannot find it. However, Kate--in case you need some light reading this weekend--I have some cites below for you.
I believe most of the evidence regarding board composition as it relates to insider vs. outside directors is equivocal. Below are two cites that both survey the evidence on this issue and add some additional evidence. There is also a study related to composition on audit committees I believe cited in the Bhagat and Black piece suggesting that such committees need an appropriate balance of inside and outside directors to be effective.
Sanjai Bhagat and Bernard Black, The Uncertain Relationship Between Board Composition and Firm Performance, 54 Bus. Law. 921 (1999); Sanjai Bhajat & Bernard Black, The Non-Correlation Between Board Independence and Long-Term Firm Performance, 27 Corp. L. 231 (2002).
There also have been some studies on composition as it relates to diversity. Some cites below. It is my understanding that the studies find some correlation (as opposed to causation) between board diversity and corporate governance issues.
Gender Diversity and Firm Value in the Journal of the Academy of Business and Economics, http://www.findarticles.com/p/articles/mi_m0OGT/is_2_5/ai_n16619655; Board Diversity and Firm Financial Performance, http://www.blackwell-synergy.com/links/doi/10.1111%2F1467-8683.00011; Catalyst study on corporate performance and gender diversity, http://www.catalystwomen.org/files/fullfinancialperformancereport.pdf; David A. Carter et.al, Corporate Governance, Board Diversity and Firm Value , http://ssrn.com/abstract_id=304499 (Social Science Research Network).
From a theoretical perspective, many authors have hypothesized that to the extent a board’s homogeneity could lead to “groupthink” and polarization, board diversity can reverse that trend thereby improving corporate governance. Some cites on that: Lynne L. Dallas, The New Managerialism and Diversity on Corporate Boards of Directors, 76 TUL. L. REV. 1363, 1403–05 (2002) Marleen A. O’Connor, The Enron Board: The Perils of Groupthink, 71 U. CIN. L. REV. 1233, 1306–08 (2003) (noting that “diversity may enhance board effectiveness”); Steven A. Ramirez, Diversity and the Boardroom, 6 STAN. J.L. BUS. & FIN. 85 (2000); Steven A. Ramirez, A Flaw in the Sarbanes-Oxley Reform: Can Diversity in the Boardroom Quell Corporate Corruption?, 77 ST. JOHN’S L. REV. 837 (2003) ; Janis Sarra, The Gender Implications of Corporate Governance Change, 1 SEATTLE J. FOR SOC. JUST. 457, 494 (2002).
6. Posted by Kate Litvak on November 10, 2006 @ 13:12 | Permalink
Vic, the Bhagat paper you are talking about has a co-author who just happened to be my spouse. I assure you, they have not found any relationship between board composition and company performance. In particular, they found that independent directors do not improve company performance. Worse yet, Bhagat and Black had no way to disentangle causation. They had an instrument, but it was a bad instrument, and there was no way to improve it. That's why the paper wasn't published in a real journal.
7. Posted by Kate Litvak on November 10, 2006 @ 13:44 | Permalink
Lisa, none of these studies even claim to establish causation! Not even close. They have no endogeneity controls of any sort, and wisely say nothing about diversity "enhancing" or "improving" anything. Notice that the authors use the correct language (“correlated”, “associated”), not the language you are attributing to them (“caused”, “enhanced”). Correlation does not equal causation.
Do you have anything that actually supports the causation claim? (Hint: I am not aware of such work, but it doesn't mean it doesn't exist...)
8. Posted by Lisa Fairfax on November 12, 2006 @ 9:05 | Permalink
Kate, as I mentioned in my post to you, the studies I cited "find some correlation (as opposed to causation) between board diversity and corporate governance issues." Hence, I did not attribute any language of causation to them. Of course by suggesting that corporate governance is enhanced by women directors, the TIAA-CREF study does speak in terms of causation and I leave you to your critique of that study and its methodology.
With regard to the larger body of empirical evidence in this area, I would say that the jury is still out. One of the critiques regarding the normative claim that board diversity enhances corporate governance was that there was no empirical evidence to support that claim. The empirical evidence to date reveals some correlation. To that end, such evidence neither proves nor disproves the causal link, thereby leaving open the possibility that some causal link can be established--and hence that the normative claim has some merit. I too am waiting for a study that can provide a more definitive answer on this possibility. Until then, again, I would say the jury is still out.
9. Posted by Kate Litvak on November 13, 2006 @ 20:52 | Permalink
No, Lisa, the jury is not out. To send the jury out, you need to produce at least some evidence rejecting the null hypothesis of no causal relationship between board composition and company performance. As we just established, you are not aware of any such evidence. I am quite sure it doesn’t exist. There are no respectable studies, zero, zilch, nada, none, supporting the claims that diversity “enhances” company performance. This line of "scholarship" is nothing but a politically correct fantasy. The null hypothesis stands undisturbed, and jury is not out. Find at least something, and we’ll talk about the jury.
I am quite surprised that you continue to mention correlation studies here. They are utterly irrelevant for our conversation. It’s possible and common to have correlation and causation pointing in the opposite directions; it’s also possible and common to have them point in the same direction; likewise, it’s possible to find correlation and no causation. There is no way to tell upfront which way their relationship will work out. These are just apples and oranges. Take the alleged (highly suspect) finding that better companies are more likely to have women directors. Inference (a): women directors improve company performance. Inference (b): women directors actually hurt company performance, but having women directors is a nice (but expensive) PR gimmick, so only the best companies can afford to pay for such scheme by willingly foregoing profitability. Inference (c): the gender of a board member is completely irrelevant for company performance, and best companies appoint women directors as a costless PR gimmick precisely because the gender of a board member is irrelevant, while less-profitable companies have more important things to worry about than meaningless PR.
The most likely inference is (d): the correlations findings are spurious – those papers don’t have sufficient controls for company- and industry-level characteristics (size, leverage, blockholdings, sales, EBIT, the industry's HHI, whether the industry is shrinking or expanding, etc) ; they are based on small and biased samples, and so forth. Given the existing state of the upscale finance literature (finding almost no correlations between any measures of corporate governance and performance in the US), the option (d) is probably right.
10. Posted by Vic on November 14, 2006 @ 8:05 | Permalink
Wrong paper, Kate. http://www.theconglomerate.org/2006/07/empirical_corpo.html