Gender Diversity in Corporate Governance and Top Management
and Financial Performance
Claude Francoeur
HEC Montréal
Réal Labelle
Chair in Governance and Forensic
Accounting
HEC Montréal
Bernard Sinclair-Desgagné
Chair in International Economics and Governance
HEC Montréal
May 2006
* Financial support from the Autorité des marchés financiers and Desjardins Securities is gratefully
acknowledged.
1
Gender Diversity in Corporate Governance and Top Management
and Financial Performance
Abstract
By requiring that a majority of board members and all audit committee members be
independent, the current reform introduces a greater degree of diversity of opinions and
interests into corporate governance and keeps management’s control over the decision
process within proper bounds. This article examines the concept of diversity by focusing
on whether and how the participation of women in the firm’s governance and their
presence among its senior management might enhance corporate performance. We build
on the agency and stakeholder theoretical frameworks (Tirole 2001; Donaldson and
Preston, 1995) to develop our main hypothesis relating gender diversity in the board
room and among senior managers to corporate performance. We use the Fama and
French (1992, 1993) valuation framework to take the level of risk into consideration
when comparing firm performances, whereas previous studies used either raw stock
returns or accounting ratios.
Results indicate that firms operating in complex environments do indeed generate
positive and significant abnormal returns, when they have a high proportion of women
officers. Although the participation of women as directors does not seem to make a
difference in this regard, firms with a high proportion of women in both their
management and governance systems generate enough value to keep up with normal
stock-market returns. Taken together, these results are compatible with the current efforts
observed in some countries and organisations towards the adoption of normative policies
for the advancement of women in business.
Key words: Corporate governance; gender diversity; stakeholders.
2
1. Introduction
Capital market participants, in particular institutional investors and ethical funds, are
paying closer attention to the governance of listed companies in pursuing both the
financial and social goals of their stakeholders. To that effect, the trend is towards greater
diversity in governance. For instance, by requiring that a majority of board members and
all audit committee members be independent, the current reform has already introduced a
greater diversity of opinions and interests into corporate governance in order to keep
management’s control over the decision process within proper bounds. The objective of
this article is to examine the concept of diversity or pluralism in corporate governance by
focusing on whether and how the participation of women in the firm’s governance and in
senior management might enhance corporate performance. We study this facet of
pluralism in governance by trying to provide answers to the following two research
questions:
- What are the characteristics or factors which lead firms to diversify their board of
directors and their management team by appointing a greater percentage of women?
- What is the impact of this diversity on the firm’s financial performance?
Our measures of women’s participation in the firm’s governance and its top-level
management are taken from the Catalyst’s1 2001-2004 annual surveys. Compared to
previous studies, our contribution is twofold. We propose a theoretical framework
relating corporate gender diversity to firm performance. We then conduct an empirical
study to test the combined effect of diversity in governance and senior management on
3
the firm’s performance. Furthermore, whereas previous studies used either raw stock
returns or accounting ratios, we apply the Fama-and-French (1992, 1993) valuation
framework to take risk level into consideration when comparing firm performance. To
develop our main hypothesis relating gender diversity on the board and in top
management to corporate performance, we build on agency and stakeholder theoretical
frameworks (Tirole 2001; Donaldson and Preston, 1995).
In line with our hypothesis, results indicate that firms operating in complex environments
which have a high proportion of women officers will generate positive and significant
abnormal returns of approximately 6 % over three years. However, no significant excess
returns are generated where women participate as directors or where their impact is
measured by a score combining their representation as both officers and directors.
Though it seems that the presence of women directors makes no difference as far as
financial performance is concerned, firms with a high proportion of women in both their
management and governance systems do generate enough value to keep up with normal
stock-market returns. Taken together, these results are compatible with the current efforts
some countries and organisations are making to adopt normative policies for the
advancement of women in business.
The remainder of this article is structured as follows. Section 2 proposes a theoretical
framework for relating pluralism and, especially, gender diversity in corporate
governance and high-level management to firm performance. Section 3 examines prior
1. Catalyst is the leading research and advisory organisation in North America working to advance women in business.
4
empirical evidence relevant to our study. Section 4 presents the data and the methodology
used. The results of the empirical analysis are presented in section 5, while section 6
concludes the paper.
2. Theoretical development and hypotheses
Over the years, the issue of gender diversity in the workforce has received a lot of
attention both in the academic literature and in the popular press, particularly in studies
addressing the “glass ceiling effect” (Farrell and Hersch 2001). This metaphor suggests a
transparent barrier which women face as they attempt to achieve promotion into the
higher levels of organizations (Li and Wearing 2004). Moreover, the presence of women
on corporate boards is considered as the “ultimate glass ceiling” by some authors (Arfken
et al., 2004). Our study relates to research that focuses on gender diversity or female
representation in board and top-management positions (Adams and Ferreira, 2004;
Kesner 1988; Bilimoria and Piderit 1994; Daily et al. 1999; Farrell and Hersch 2001) and
on whether it can enhance corporate performance (Catalyst 2004; Carter et al., 2003;
Adler 2001).
Despite the topic’s latent interest , there are relatively few empirical academic studies
linking gender diversity in corporate governance or in top management to firm
performance. Yet , several studies have examined how other board characteristics may be
5
related to performance.2 Most of these begin with the assumption that the director’s
effectiveness is a function of the board's independence from management. According to
Fields and Keys (2003: 13), the issue of diversity may be linked to this more general
issue of independent outside directors, since women and minorities are found to be more
likely outsiders (Carter et al. 2003).
From their review of these studies on boards of directors, Hermalin and Weisbach (2003)
report that a number of regularities have emerged: notably, that board composition, a
proxy for independence, does not seem to predict corporate performance, whereas board
size is negatively related to performance. The unobservable nature of board independence
may make it difficult to link it to performance, but this is not the case for the gender
issue. One reason for this lack of attention to the gender issue (and the difficulty in
establishing its clear-cut link with performance) may be because little formal theory has
been developed on boards of directors, let alone on their gender diversity. Agency theory
does provide an explanation for the existence of boards of directors as a response to
agency problems but nothing with regard to a clear-cut prediction governing the
relationship between gender diversity and performance.
The objective of this paper is to contribute to the development and exploration of a
theoretical framework aimed at linking gender diversity and performance . We use the
agency and stakeholder theoretical frameworks (Tirole 2001; Donaldson and Preston,
1995) to develop our main hypothesis relating gender diversity in the board room and top
2
See, for instance, Bhagat and Black, 2002; Hermalin and Weisbach, 1991; Kaplan and Reishus, 1990;
6
management to corporate performance.
Agency theory
In his Presidential address for the Econometric Society, Tirole (2001, p. 2) defines a
“good” corporate governance structure as “one that selects the most able managers and
makes them accountable to investors” (emphasis added). Up to now, however, the best
means of implementing these desiderata are still being sought.3
As shown in the previous section, one possible means which has recently shown up on
the agendas of academic researchers and corporate board members is increased gender
diversity. Few women currently sit on the boards of major corporations. Yet, the
preliminary empirical evidence, reviewed in section 3, suggests that better financial
performance might result from increasing the female presence on corporate boards and
among senior management.
A common rationale for these findings would go as follows: women (like external
stakeholders, ethnic minorities, and foreigners) often bring a fresh perspective on
complex issues and this can help correct informational biases in formulating strategies
and solving problems. While intuitively plausible and well-grounded in cognitive
psychology and decision theory, this explanation largely overlooks the concrete tasks
Klein, 1998.
3
Like most finance, accounting and economics scholars, we center here on how to foster shareholder
value rather than on whether such an objective is legitimate.
7
mentioned in the above quote. Before it gives advice on strategy, for instance, a corporate
board must first appoint the right person as CEO. Special attention will then be paid to
the leadership style that seems most appropriate in the actual business landscape. An
autocratic leader might do well in pulling the firm out of a financial crisis, but a more
democratic leader will better stimulate creativity and intrapreneurship in highly
competitive, innovation-driven industries [Rotemberg and Saloner 1993]. However, if
preferences for leadership styles are more influenced by gender than by economic
considerations, a diverse board might well end up making the wrong appointment,
thereby providing an untimely leadership style (or no leadership at all).
Corporate boards must also motivate and monitor the acting CEO. In appraising the
latter’s work, female directors might put forward the interests of employees and other
stakeholders on issues such as non-financial disclosures, career management, and job
discrimination. But saddling the CEO with too many (often conflicting ) requirements
might well blur the bottom line pursued and dilute incentives, as the more recent
developments in agency theory currently show (see, for instance, Dixit 1996 and Heath
and Norman 2004).
From a theoretical standpoint, when one considers the overall impact of gender diversity
on the various duties being assumed by a corporate board, it would seem impossible to
predict that greater female participation will improve (or impair) corporate governance
and, as an indirect consequence, corporate performance. Different factors seem to pull in
different directions. To find out which of the factors identified are likely to prevail, we
8
must now turn to empirical research. This is what we shall endeavour to do in section 4.
Stakeholder theory
Before we turn to the empirical portion of this paper, let’s briefly examine what the
stakeholder theory says on our research question. Though the agency theory framework
has indeed been very often used to study phenomena concerning corporate governance
and top management, firms have, in recent years, been increasingly pressured by
shareholder activists, large institutional investors (Fields and Keys, 2003: 12), and
politicians to appoint women as directors or officers. The assumption is that “greater
diversity of the boards of directors should lead to less insular decision making processes
and greater openness to change” (Westphal & Milton 2000).
In other words, despite what we noted in the previous subsection as being the conceptual
difficulty in establishing a clear link between diversity and performance, there do seem to
be strong normative and business reasons for diversity. In this subsection, we consider
stakeholder theory to explore them. Stakeholder theory has been advanced and justified
in management literature on the basis of its descriptive accuracy, instrumental power, and
normative validity (Donaldson and Preston, 1995: 65).
Regarding the first perspective, many studies conducted— notably by the Conference
Board (2005) and Catalyst (2001-2004)—have confirmed the descriptive accuracy of
stakeholder theory by longitudinally tracking the presence of women on corporate boards
9
or among top managers.
To explore its research question, our study relies mainly on the instrumental facet of
stakeholder theory. This facet establishes a framework for examining the assumed
relationship or connection between the achievement of various corporate performance
goals and gender diversity. The principal focus of interest here is the proposition that
corporations practicing stakeholder management—in our case, giving greater voice to
women—will, other things being equal, be relatively successful.
Dallas (2002) surveys some psychological research that attempts to explore the effects of
group member characteristics and heterogeneity, such as gender diversity, on group
decision-making. To summarize this line of research: It seems that, in today’s complex
and rapidly changing business environment, when it comes to engendering better quality
decisions, the advantages related to the knowledge, perspective, creativity, and judgment
brought forward by heterogeneous groups may be superior to those related to the
smoother communication and coordination expected from homogeneous groups. This
superiority is achieved by taking a wider range of alternatives and consequences into
consideration.
We also grounded the above instrumental perspective of stakeholder theory on its
normative aspect. We interpret this normative perspective as being less stringent than the
instrumental one, in the sense that diversity is a sensible objective even if it does not
necessarily lead to improved performance. From that perspective, even if no relationship
10
(neither negative nor positive) were found between diversity and performance, diversity
would still be a good policy to pursue. This perspective has some empirical support. For
instance, Dallas (2002) observes that issues such as family life and flexible work
arrangements are given greater prominence in companies that attract both female
executives and female board members. Norway has adopted a law with regard to the
promotion of women in business and, more recently, one of Canada’s provincial
legislatures has adopted a resolution to gradually increase to 50 % the proportion of
women sitting on the boards of public institutions (Audet 2006).
Many reasons for having women on boards and among top managers simply make good
business sense and support instrumental as well as normative perspectives, reasons such
as drawing from a wider range of available intellectual capital, catering to women’s
purchasing power, role modeling and signalling to employees—particularly current
female managers and potential recruits.
3. Prior empirical evidence
Empirical research on an assumed relationship between gender diversity in governance
and top management and corporate financial performance is sparse. This is especially
true if studies dealing with other aspects of diversity such as racial, ethnic, cultural, etc.
are omitted. In this section, we review the few studies that are directly relevant to our
research question.
11
Using a sample of 200 large US firms, Shrader et al. (1997) fail to find any significant
relation between the percentage of women in the upper echelons of management and firm
performance. As for the participation of women on boards, they find a negative impact on
performance. The authors note that the low percentages of women among top managers
or on boards —4.5% and 8% respectively— could impair the validity of their findings.
Performance is measured using accounting data such as return on assets (ROA), return on
sales (ROS), return on investments (ROI), and return on equity (ROE).
Adler (2001) shows a strong correlation between women-friendly firms and high
profitability. The sample is composed of 25 Fortune 500 firms showing a strong record of
participation of women in the executive suite. Three accounting measurements of
operational performance are used (ROS, ROA and ROE).
Carter et al. (2003) find a positive relationship between board diversity (in terms of
women and minorities) and firm value. Using a sample of 638 Fortune 1000 firms, the
results of this study suggest that a higher percentage of women and minorities on the
board of directors can increase firm value, as proxied by Tobin’s Q. The study also shows
that the proportion of women on boards is a significant determinant of the fraction of
minority directors on boards. This research does not provide a clear-cut conclusion on the
effect of the greater participation of women alone on the firm’s value.
12
Adams and Ferreira (2004) document that boards of directors tend to be more
homogeneous when firms operate in riskier environments. On the basis that social
homogeneity breeds trust, an argument put forward by Kanter (1977), this study shows
that diverse boards receive additional compensation to palliate the decrease in
homogeneity. This, in turn, may reduce the firm’s performance and decrease its value. On
the other hand, “firms with more diverse boards hold more board meetings” and “female
directors have fewer attendance problems at board meetings.” These later findings
militate for the greater effectiveness of diverse boards. The net effect of gender diversity
remains unclear.
In 2004, Catalyst, a well-known organization often cited for its research on the place of
women in business, has conducted a study on the presumed connection between gender
diversity and financial performance. Using a sample of 353 Fortune 500 companies from
1996 to 2000, the research shows that gender diversity has a positive impact on financial
performance, as measured by ROE and raw stock returns. Their measure of diversity is
based solely on the participation of women as corporate officers. They conclude that the
firms belonging to the top quartiles in terms of diversity achieve a higher financial
performance than their counterparts in the lower quartile.
13
3. Data and methodology
In this section, we present the data and methodology used, while underlining the steps we
took to try to improve on previous research. Our measures of women’s participation in
the firm’s governance and top management are respectively collected from the 2001 and
2003 Catalyst censuses of women board directors and from the 2002 and 2004 Catalyst
censuses of women officers in the Financial Post’s list of the 500 largest Canadian firms.
For the purpose of our study, we have combined the annual Catalyst data by computing
the average representation of women over the two-year period. If a firm with 1 woman
officer in one year has none in the other, it is accounted for as having an average of 0.5
women officers. When a company’s statistic is only available for one year, we have used
it. Given the stability of the statistics as outlined by Catalyst in its latest census (2005, p.
1), these methodological choices should not bias our results.
In 2005, the story of women in corporate governance in Canada continued to
be the same as in previous years: slow growth. Women held 12.0 percent of
all board seats among the FP500 companies, up from 11.2 percent in 2003—
only a 0.8 point increase in two years, and consistent with the trends we have
witnessed in the United States and Canada since we started counting women
board directors.
We have also computed a total percentage of women directors and officers to test the
combined effect of diversity in governance and top management on the firm’s
performance. This proportion is the sum of the number of women directors and the
number of women officers divided by the total number of positions.
Furthermore, to access the degree of variation in female representation across firms and
14
to test its hypothesised positive relation with performance, we have split the sample in
three. This grouping allows us to compare firms in the first third— showing a low
proportion of women— to those in the last third— exhibiting a high proportion of
women. Indeed, our subsequent univariate and multivariate analyses will use this
grouping to test for significant differences between the high and low groups as far as
their performance, risk, and complexity are concerned. As most of the data used in this
study are not normally distributed or skewed to the right, we shall use non-parametric
tests to compare group means.
As previous studies have underlined the fact that diversity should help to deal with risky
and complex situations, we have developed a number of measurements to account for
these factors. Our proxies for risk and complexity are the firm’s beta, the market-to-book
ratio and the standard deviation of the analyst’s error forecasts. The Market-to-Book
ratios are taken from Stock Guide which publishes monthly accounting data and financial
ratios taken from the latest financial reports of Canadian public firms. Standard
deviations of analyst’s error forecasts are obtained from the Institutional Brokers‘
Estimate System (IBES). Monthly average betas were computed from data obtained from
the Toronto Stock Exchange (TSX) data bank.
The few studies (Catalyst 2004, Carter et al. 2003, Adler 2001 and Shrader et al. 1997)
that have examined the effect of diversity on long term-financial performance have used
a simplistic approach to measure returns. Indeed, in measuring returns, they took no
account of risk, which is a fundamental factor in finance when attempting to compare the
15
financial performance of firms. Following theses studies, we first compute monthly
average gross returns for the period from January 2002 to December 2004. Then, to
control for risk, we use the three-factor Fama/ French (1992, 1993) valuation model. This
model is deemed to explain the firm’s return by its beta, its size and its book-to-market
ratio. If it does not entirely explain the firm’s financial performance, i. e. if the alpha
coefficient of the linear model is significant (to be interpreted as an abnormal return) this
indicates an omitted variable. Running it separately for the firms with a low and a high
proportion of women, we will reject the null hypothesis of no relation between the
representation of women and firm performance if the alpha is significant in the case of
firms with the highest proportion of women.
4. Results and analysis
In this section, we first present statistics on the extent of the difference existing between
firms with a low as opposed to those with a high proportion of women as officers,
directors or both. We then present and interpret our univariate analyses of the factors
which are presumed to characterize firms that diversify their board of directors and their
senior management team by appointing women. Finally, we use the three-factor
Fama/French (1992, 1993) model to examine the presumed relationship of this gender
diversity with the firm’s financial performance.
Descriptive statistics and univariate analysis
Table 1 provides descriptive statistics on the average representation of women officers
(panel A) and directors (panel B) as well as on the combined measure of the proportion
16
of women officers and directors (panel C) in FP500 firms. It shows that, on average,
women represent 10.8 % of the officers in FP500 firms, while they only hold 7 % of the
board seats. Our combined measure shows an average of 9.1 % participation. The JarqueBera tests reject the nul hypothesis of normality of the distributions of women
representation(s)
Table 2 shows the descriptive statistics for the variables measuring the firms risk and
complexity. Our proxy for complexity stands at an average of 1.77 when measured as the
firm’s Market-to-Book ratio and at 0.04 when measured as the forecast’s standard
deviation. In general, the Jarque-Bera tests reject the nul hypothesis of normality of the
distributions of our measures of risk and complexity.
Table 3 provides the firm’s average monthly returns for the period 2002 to 2004. Firms
are grouped according to women’s participation as directors, officers or both. The
average monthly returns are 1.43 %, 1.21 %, and 1.25% respectively. The slight
difference in the number of firms between the three subcategories is due to the data
available from the Catalyst censuses and the availability of the returns from the TSX data
bank. The Jarque-Bera tests reject the nul hypothesis of normality for the distributions of
average monthly returns.
Table 4 compares the average monthly returns of firms with a low proportion of women
officers to those with a high proportion (panel A). Then, the two groups of firms are
again compared after being separated on both sides of the median between low and high
17
beta to take risk and complexity into consideration as required in theory. The same
comparison is done for low and high market-to-book (panel B) and forecast standard
deviation (panel C). Using non-parametric tests, none of the difference is statistically
significant. Tables 5 and 6 repeat those comparisons for firms with women as directors
and for firms with women as both officers and directors respectively. Again, generally, at
the conventional level of 5%, there is no significant difference between the two groups
when risk and complexity are taken into consideration.
In summary, our non-parametric univariate analysis does not exhibit any statistically
significant difference between the firms with low and high gender diversity even when
taking risk and complexity into consideration. Therefore, the instrumental perspective of
the stakeholder theory does not seem to be supported. But the results seem to be
supportive of the normative perspective, which makes gender diversity a good policy to
pursue even if no significant negative relationship is found between diversity and
performance. Given these weak results obtained for our main hypothesis, let’s now
proceed with our multivariate analysis.
Multivariate analysis
In this section, we use a multivariate approach to determine whether gender diversity is
associated with the firm’s financial performance. Our analysis is based on the threefactor Fama/French multivariate model (1992, 1993). This model is widely used in the
literature to estimate a firm’s expected return as a function of its beta, its size and its
book-to-market ratio (BM). Running it separately for the firms with a low and a high
proportion of women, the null hypothesis of no relation between the representation of
18
women and firm performance will be rejected if alpha, representing abnormal returns, is
significant in the case of firms with the highest proportion of women.
R p t − R f t = a p + b p (R m t − R f t ) + s p SMB t + h p HML t
where R p t − R f t is the monthly excess return of the portfolio of our sample firms over
the risk free rate. R m t − R f t is the excess return required by the market over the risk free
rate, as used in the capital asset pricing model (CAPM). SMB (Small minus Big) is the
excess return required for investing in small firms rather than in big firms and HML
(High minus Low) is the excess return required for value firms (high book-to-market
ratios) as opposed to growth firms.
The portfolio and market returns are calculated on an equally weighted basis. The
intercept “ a ” indicates the monthly average abnormal return of our sample. We use
weighted least squares regressions to control for the heteroskedacity potentially induced
by the fact that the number of firms in our monthly portfolios fluctuates over time. Our
weights are the reciprocal of the square root of the number of firms in each month.
In order to account for size and book-to-market peculiarities in Canada, b, s and h
coefficients were estimated using all Toronto Stock Exchange (TSX) returns for the
period from 1990 to 2004. To compute SMB, following Fama and French (1992, 1993),
we sorted all stocks into 6 portfolios and ranked them based on their size and BM ratios.
The stocks were subsequently sorted into two size groups and three BM subgroups. Firms
above median size were designated “big” and firms below median, “small.” Firms in the
bottom 30% in terms of BM ratio were designated “low” and those in the top 30% were
19
designated “high.” The SMB factor represents the average excess return of small firms
over big firms. The HML factor represents the average excess return of value firms (high
BM ratios) over glamour or growth firms (low BM ratios).
In order to validate our parameters, for every month from January 1990 to December
2004, we formed 25 portfolios using the size and book-to-market ratios of all TSX firms
for which we found values in the Stock Guide database. For every month, we ranked and
sorted all firms into five groups based on size and into five subgroups based on BM ratio.
For every month, we ran 25 regressions of the Fama and French model. We used 91-day
Canadian Government Treasury Bills as a proxy for the risk free rate (Rf). The market
return is the equally weighted value of all stocks quoted on the TSX. As expected by the
Fama and French model, our results show that the SMB and HML factors used are
significant drivers of excess returns for Canadian firms.4
Table 7 shows that firms operating in complex environments, as measured by high betas,
high market-to-book ratios or analysts’ forecasts standard deviation, generate positive
and significant abnormal returns when they have a high proportion of women officers.
From an economic perspective, the observed excess return is in the order of 6 % over
three years (0.17 % compounded over 36 months). This result is robust whatever
measurement of complexity used and supports the instrumental view of stakeholder
theory.
4
These analyses are available upon request
20
Tables 8 and 9 show that women acting as directors or as measured by a score combining
the representation of both officers and directors do not generate significant excess
returns. Those results indicate that although the participation of women as director does
not seem to make a difference as far as financial performance is concerned, firms with a
high proportion of women are able to generate enough value to keep up with normal
stock-market returns.
5. Conclusions
Tirole (2001: 2) defines a “good” corporate governance structure as “one that selects the
most able managers and makes them accountable to investors.”.However, to date. the
best means of implementing these desiderata are still being sought. One possible means
which has recently shown up on the agendas of academic researchers and corporate board
members is increased gender diversity. Few women currently sit on the boards of major
corporations. Yet, the preliminary empirical evidence, which we reviewed in section 3 of
this article, suggests that better financial performance might result from increasing the
female presence in these bodies.
This article’s contribution to this line of research is twofold. We propose a theoretical
framework for a relationship between gender diversity and firm performance. Then, we
empirically test the combined effect of diversity in governance and top management on
the firm’s performance. We use the Fama and French (1992, 1993) valuation framework
to take the level of risk into consideration when comparing firm performances, while
21
previous studies either use raw stock returns or accounting ratios. We build on the agency
and stakeholder theoretical frameworks to develop our main hypothesis of a relation
between gender diversity in the board room and top management and corporate
performance.
Results indicate that firms operating in complex environments indeed generate positive
and significant abnormal returns of approximately 6 % over three years, when they have
a high proportion of women officers. This result is in line with our hypothesis. However,
women participation as directors or as measured by a score combining their
representation as both officers and directors does not generate significant excess returns.
Although the participation of women as directors does not seem to make a difference as
far as financial performance is concerned, firms with a high proportion of women in both
their management and governance systems generate enough value to keep up with normal
stock-market returns. Taken together, these results are compatible with the current efforts
observed in some countries and organisations which seek the prescription of normative
policies for the advancement of women in business.
22
Table 1
Representation of Women
For the period 2001 to 2004
All firms
A
Officers
2002 and 2004
B
Directors
2001-2003
C
Directors and Officers
2001-2004
Number of firms
Mean (%)
Median (%)
Maximum (%)
Minimum (%)
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
Number of firms
Mean (%)
Median (%)
Maximum (%)
Minimum (%)
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
Number of firms
Mean (%)
Median (%)
Maximum (%)
Minimum (%)
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
234
10.7808
9.1000
50.0000
0.0000
10.2617
1.0436
4.2693
58.1841
0.0000
229
7.0201
4.8000
50.0000
0.0000
8.1210
1.2489
5.5062
119.4605
0.0000
219
9.0639
8.1000
33.3000
0.0000
7.2113
0.6508
2.8818
15.5871
0.0004
Low Percentage High Percentage
firms
firms
78
0.4910
0.0000
4.8000
0.0000
1.3137
2.3867
6.9853
125.6693
0.0000
105
0.0000
0.0000
0.0000
0.0000
0.0000
72
1.5361
0.0000
4.9000
0.0000
1.8179
0.4958
1.5474
9.2791
0.0097
74
22.8743
20.0000
50.0000
14.8000
7.8405
1.7065
5.9451
62.6620
0.0000
74
16.7392
15.4000
50.0000
10.5000
6.0358
2.6921
14.0616
466.6560
0.0000
73
17.4836
16.0000
33.3000
11.8000
4.6306
0.9922
3.7296
13.5974
0.0011
23
Table 2
Risk and complexity of the sample firms
For the period 2002 to 2004
All
Three year average monthly
Beta
Market-to-Book
Forecast Std Deviation
Number of firms
Mean
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
Number of firms
Mean ratio
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
Number of firms
Mean
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
179
0.6775
0.5505
2.9266
0.0036
0.5500
1.6758
6.2419
162.1679
0.0000
187
1.7714
1.5062
12.8900
0.0306
1.3556
3.7509
27.2250
5011.0340
0.0000
134
0.0401
0.0219
0.7978
0.0000
0.0791
7.0025
64.6301
22302.1100
0.0000
Low
High
Risk/Complexit Risk/Complexit
y
y
82
0.2714
0.2599
0.5217
0.0036
0.1468
0.0217
1.9748
3.5973
0.1655
95
0.9664
0.9997
1.5100
0.0306
0.3505
-0.3838
2.2749
4.4133
0.1101
70
0.0097
0.0091
0.0229
0.0000
0.0083
0.1500
1.4767
7.0303
0.0297
97
1.0207
0.8606
2.9266
0.5258
0.5318
1.7562
5.5160
75.4464
0.0000
92
2.6027
2.1587
12.8900
1.5421
1.5013
4.0747
25.9614
2275.6200
0.0000
64
0.0733
0.0420
0.7978
0.0230
0.1049
5.4241
37.0749
3410.0880
0.0000
24
Table 3
Average monthly returns
Full sample of firms with women
For the period 2002 to 2004
Officers
Directors
N
Mean
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
N
Mean
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
Directors & Officers
N
Mean
Median
Maximum
Minimum
Std Deviation
Skewness
Kurtosis
Jarque-Bera
Probability
233
0.0143
0.0137
0.1321
-0.1610
0.0316
-1.1504
10.7697
637.4724
0.0000
230
0.0121
0.0126
0.1321
-0.2044
0.0354
-1.9278
13.5376
1206.6110
0.0000
222
0.0125
0.0131
0.1321
-0.2044
0.0348
-1.9345
14.3560
1331.3240
0.0000
25
Table 4
Average Monthly Returns – Women Officers
For the period 2002 to 2004
A
Women Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
B
N
78
74
Median
0.0127
0.0148
Value
0.4055
0.4055
0.1053
0.0263
0.1659
0.1659
0.1098
P-Value
0.6851
0.6851
0.7455
0.8711
0.6838
0.6838
0.7404
Women Officers
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
C
All
All
N
75
69
Median
0.0126
0.0149
Value
0.6618
0.6618
0.2504
0.1113
0.4407
0.4407
0.3107
P-Value
0.5081
0.5081
0.6168
0.7387
0.5068
0.5068
0.5773
Women Officers
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
All
N
51
54
Median
0.0119
0.0153
Value
0.7149
0.7149
0.0101
0.0090
0.5156
0.5156
0.5222
P-Value
0.4747
0.4747
0.9200
0.9244
0.4727
0.4727
0.4699
Low Beta
N
Median
39
0.0114
37
0.0135
Value
0.1611
0.1611
0.0527
0.0000
0.0276
0.0276
0.1039
P-Value
0.8720
0.8720
0.8185
1.0000
0.8679
0.8679
0.7472
Low Market-to-Book
N
Median
41
0.0125
35
0.0155
Value
0.3231
0.3231
0.4767
0.2118
0.1078
0.1078
0.0186
P-Value
0.7467
0.7467
0.4899
0.6453
0.7427
0.7427
0.8915
Low Forecast SD
N
Median
24
0.0128
31
0.0149
Value
0.4158
0.4158
0.0141
0.0235
0.1800
0.1800
0.2885
P-Value
0.6776
0.6776
0.9055
0.8782
0.6714
0.6714
0.5912
High Beta
N
Median
39
0.0190
37
0.0205
Value
0.6339
0.6339
0.0527
0.0000
0.4085
0.4085
0.4386
P-Value
0.5261
0.5261
0.8185
1.0000
0.5227
0.5227
0.5078
High Market-to-Book
N
Median
34
0.0142
34
0.0148
Value
0.3434
0.3434
0.0000
0.0588
0.1222
0.1222
0.3498
P-Value
0.7313
0.7313
1.0000
0.8084
0.7267
0.7267
0.5543
High Forecast SD
N
Median
27
0.0119
23
0.0157
Value
0.5061
0.5061
0.0805
0.0000
0.2661
0.2661
0.1670
P-Value
0.6128
0.6128
0.7766
1.0000
0.6060
0.6060
0.6828
26
Table 5
Average Monthly Returns – Women Directors
For the period 2002 to 2004
A
Women Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
B
N
102
74
Median
0.0165
0.0110
Value
1.6064
1.6064
0.8394
0.5829
2.5853
2.5853
2.4312
P-Value
0.1082
0.1082
0.3596
0.4452
0.1079
0.1079
0.1189
Women Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
C
All
All
N
102
70
Median
0.0165
0.0110
Value
1.7283
1.7283
0.8672
0.6022
2.9924
2.9924
2.7885
P-Value
0.0839
0.0839
0.3517
0.4377
0.0837
0.0837
0.0949
Women Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
van der Waerden
All
N
66
50
Median
0.0165
0.0114
Value
0.8223
0.8223
0.5624
0.3164
0.6808
0.6808
0.4127
P-Value
0.4109
0.4109
0.4533
0.5738
0.4093
0.4093
0.5206
Low Beta
N
Median
39
0.0156
43
0.0111
Value
1.3231
1.3231
0.0489
0.0000
1.7630
1.7630
1.7984
P-Value
0.1858
0.1858
0.8250
1.0000
0.1843
0.1843
0.1799
Low Market-to-Book
N
Median
61
0.0203
28
0.0128
Value
0.9675
0.9675
0.7078
0.3758
0.9446
0.9446
0.8989
P-Value
0.3333
0.3333
0.4002
0.5399
0.3311
0.3311
0.3431
Low Forecast SD
N
Median
38
0.0100
21
0.0106
Value
0.1187
0.1187
0.1360
0.0094
0.0160
0.0160
0.0800
P-Value
0.9055
0.9055
0.7123
0.9229
0.8992
0.8992
0.7773
High Beta
N
Median
63
0.0170
31
0.0109
Value
1.0857
1.0857
1.2033
0.7701
1.1875
1.1875
0.9355
P-Value
0.2776
0.2776
0.2727
0.3802
0.2758
0.2758
0.3334
High Market-to-Book
N
Median
41
0.0114
42
0.0107
Value
0.9427
0.9427
0.1076
0.0118
0.8973
0.8973
0.8075
P-Value
0.3458
0.3458
0.7429
0.9136
0.3435
0.3435
0.3689
High Forecast SD
N
Median
28
0.0189
29
0.0117
Value
1.8117
1.8117
2.9587
2.1173
3.3114
3.3114
2.9278
P-Value
0.0700
0.0700
0.0854
0.1456
0.0688
0.0688
0.0871
27
Table 6
Average Monthly Returns – Women Officers & Directors
For the period 2002 to 2004
A
Women Officers & Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
Van der Waerden
B
All
N
72
73
Median
0.0165
0.0137
Value
1.0282
1.0282
0.5578
0.3373
1.0612
1.0612
1.1568
P-Value
0.3039
0.3039
0.4551
0.5614
0.3029
0.3029
0.2821
Women Officers & Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
Van der Waerden
C
All
N
74
68
Median
0.0165
0.0139
Value
0.9985
0.9985
0.4515
0.2540
1.0010
1.0010
1.1179
P-Value
0.3181
0.3181
0.5016
0.6143
0.3171
0.3171
0.2904
Women Officers & Directors
Low Percentage of Women
High Percentage of Women
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis
Kruskal-Wallis (tie-adj.)
Van der Waerden
All
N
48
52
Median
0.0179
0.0136
Value
0.6589
0.6589
0.6410
0.3606
0.4387
0.4387
0.2895
P-Value
0.5100
0.5100
0.4233
0.5482
0.5078
0.5078
0.5906
Low Beta
N
Median
28
0.0126
37
0.0146
Value
0.5100
0.5100
0.1545
0.0203
0.2669
0.2669
0.4536
P-Value
0.6100
0.6100
0.6942
0.8866
0.6054
0.6054
0.5006
Low Market-to-Book
N
Median
45
0.0217
29
0.0155
Value
1.2180
1.2180
0.5103
0.2268
1.4970
1.4970
1.7284
P-Value
0.2232
0.2232
0.4750
0.6339
0.2211
0.2211
0.1886
Low Forecast SD
N
Median
23
0.0170
27
0.0114
Value
0.1363
0.1363
0.0805
0.0000
0.0213
0.0213
0.0003
P-Value
0.8916
0.8916
0.7766
1.0000
0.8839
0.8839
0.9858
High Beta
N
Median
44
0.0181
36
0.0115
Value
0.6480
0.6480
1.8182
1.2626
0.4261
0.4261
0.2739
P-Value
0.5170
0.5170
0.1775
0.2612
0.5139
0.5139
0.6007
High Market-to-Book
N
Median
29
0.0114
39
0.0135
Value
0.2232
0.2232
0.0601
0.0000
0.0526
0.0526
0.2385
P-Value
0.8234
0.8234
0.8063
1.0000
0.8186
0.8186
0.6253
High Forecast SD
N
Median
25
0.0189
25
0.0137
Value
0.9119
0.9119
0.7200
0.3200
0.8494
0.8494
0.7563
P-Value
0.3618
0.3618
0.3961
0.5716
0.3567
0.3567
0.3845
28
Table 7
Fama & French Three Factor Model – Women Officers
For the period 2002 to 2004
Women Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
All
Low Beta
High Beta
0.0001
0.1143
0.9098
0.6731
32
0.0001
0.1858
0.8539
0.6105
33
0.0001
0.0995
0.9215
0.5890
32
0.0006
1.2020
0.2388
0.7428
34
-0.0001
-0.0587
0.9535
0.3843
34
0.0017
1.8215
0.0785**
0.7853
34
All
Low Market-to-Book
High Market-to-Book
0.0000
0.0416
0.9671
0.6503
32
-0.0010
-0.9720
0.3391
0.5416
33
0.0007
1.1299
0.2678
0.7013
33
0.0006
1.2219
0.2312
0.7471
34
0.0000
-0.0037
0.9971
0.5455
34
0.0016
2.7891
0.0091***
0.8092
34
All
Low Standard
Deviation of Forecast
High Standard
Deviation of Forecast
0.0003
0.4749
0.6384
0.6603
33
0.0005
0.6685
0.5093
0.5884
32
0.0004
0.2233
0.8248
0.5710
34
0.0008
1.6228
0.1151
0.7608
34
0.0011
1.3597
0.1844
0.4575
33
0.0019
1.8073
0.0807**
0.8028
34
***, **, * Significantly different from zero at the 1, 5, and 10 percent levels, respectively.
29
Table 8
Fama & French Three Factor Model – Women Directors
For the period 2002 to 2004
Women Directors
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Directors
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Directors
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
All
Low Beta
High Beta
0.0000
-0.0872
0.9311
0.7544
34
0.0001
0.1460
0.8849
0.5869
34
-0.0001
-0.1771
0.8606
0.7079
34
0.0000
-0.0557
0.9560
0.7372
34
0.0003
0.6114
0.5455
0.5367
34
-0.0006
-0.7079
0.4845
0.7252
34
All
Low Market-to-Book
High Market-to-Book
0.0001
0.1819
0.8569
0.7217
33
-0.0001
-0.1273
0.8995
0.6492
34
0.0000
0.0582
0.9540
0.7930
34
0.0000
-0.0557
0.9560
0.7372
34
0.0004
0.3353
0.7398
0.5343
33
0.0001
0.1680
0.8677
0.6367
34
All
Low Standard Deviation
of Forecast
High Standard Deviation
of Forecast
-0.0001
-0.1465
0.8845
0.7684
34
-0.0006
-0.4814
0.6337
0.6588
34
0.0012
1.2834
0.2095
0.6345
33
0.0003
0.7198
0.4774
0.7065
33
-0.0005
-0.4596
0.6491
0.5777
34
0.0008
1.2275
0.2295
0.6214
33
30
Table 9
Fama & French Three Factor Model – Women Directors and Officers
For the period 2002 to 2004
Women Directors & Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Directors & Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
Women Directors & Officers
Low Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
High Percentage of Women
Alpha
t-Statistic
P-Value
Adj. R Square
Nb of Months
All
Low Beta
High Beta
0.0003
0.4111
0.6841
0.6124
32
0.0003
0.3691
0.7148
0.5561
33
0.0000
0.0462
0.9634
0.6331
34
0.0004
1.1797
0.2477
0.7983
33
0.0002
0.2450
0.8081
0.4698
34
0.0006
0.7258
0.4736
0.8007
34
All
Low Market-to-Book
High Market-to-Book
0.0002
0.3902
0.6992
0.6872
33
0.0001
0.1053
0.9168
0.6552
34
-0.0003
-0.3184
0.7524
0.7127
34
0.0004
1.0712
0.2929
0.7643
33
0.0005
0.4596
0.6492
0.5717
33
0.0008
1.6034
0.1193
0.8103
34
All
Low Standard Deviation
of Forecast
High Standard Deviation
of Forecast
0.0000
-0.0507
0.9599
0.6704
33
-0.0007
-0.3471
0.7310
0.5029
33
0.0009
0.7899
0.4360
0.5668
33
0.0005
1.3827
0.1773
0.8150
33
0.0004
0.5039
0.6180
0.7310
34
0.0011
1.4495
0.1579
0.6654
33
31
References
Adams, R. B. and D. Ferreira: 2004, “Gender Diversity in the Boardroom”, European
Corporate Governance Institute, Finance Working paper # 57, 30 p.
Adler, R.D.: 2001, “Women in the Executive Suite Correlate to High Profits”, Glass
ceiling Research Center, http://glass-ceiling.com/InTheNewsFolder/HBRArticlePrintablePage.html.
Arfken, D. E., Bellar, S. L. and M. M. Helms: 2004. “The Ultimate Glass Ceiling
Revisited: The Presence of Women on Corporate Boards”. Journal of Business Ethics 50,
177–186.
Audet, M.: 2006. “To modernize the Public Corporations Governance”. Statement of
policy, 33 p.
Bilimoria, D. and S. K. Piderit: 1994, “Board committee membership: Effects of
sexbased bias”, Academy of Management Journal 37, 1453-1477.
Bhagat, S. and B. Black: 2002, “The non-correlation between board independence and longterm firm performance”, Journal of Corporation Law 27, 2 (Winter), 231-273.
Carter, D. A., B. J. Simkins and W. G. Simpson: 2003, The Financial Review 38, 33-53.
Catalyst: 2004, “The Bottom Line: Connecting Corporate Performance and Gender
Diversity”, Catalist Publication Code D58, New York, 28 p.
Daily, C. M., C. S. Trevis and D. R. Dalton: 1999, “A decade of corporate women: Some
progress in the boardroom, none in the executive suite”, Strategic Management Journal
20, 93-99.
Dallas, L. L.: 2002, “The New Managerialism and Diversity on Corporate Boards of
Directors”, Public Law and Legal Theory Working Paper 38 (Spring), University of San
Diego School of Law (http://ssrn.com/abstract_id=313425), 30 p.
32
Dixit, A. K.: 1996, “The Making of Economic Policy: A Transaction-Cost Politics
Perspective”, Cambridge (MA: MIT Press).
Donaldson, T. and L. Preston: 1995, “The stakeholder theory of the corporation:
Concepts, evidence, and implications”, The Academy of Management Review 20 (1), 65.
Fama, E.F. and K.R. French: 1992, “The Cross-Section of Expected Stock Returns”, The
Journal of Finance 47 (2), 427-465.
Fama, E.F. and K.R. French: 1993, “Common Risk Factors in the Returns on Stocks and
Bonds”, Journal of Financial Economics 33, 3-56.
Farrell, K. A. and P. L. Hersch: 2001, “Additions to Corporate Boards: Does Gender
Matter?”, Working paper, 30 p.
Fields, M. A. and P. Y. Keys: 2003. “The Emergence of Corporate Governance from
Wall St.: Outside Directors, Board Diversity, Earnings Management, and Managerial
Incentives to Bear Risk”, The Financial Review 38, 1-24.
Heath, J. and N. Wayne: 2004, “Stakeholder Theory, Corporate Governance and Public
Management: What can the History of State-Run Enterprises Teach us in the Post-Enron
era?”, Journal of Business Ethics 53, 247-265.
Hermalin, B. and M. Weisbach: 2003. “Boards of directors as an endogenously
determined institution: A survey of the economic literature”, Economic Policy Review Federal Reserve Bank of New York 9 (1), 7-26.
Hermalin, B.and M. Weisbach: 1991, “The Effects of Board Composition and Direct
Incentives on Firm Performance”, Financial Management 20 (4), 101-12.
Kanter, R. M.: 1977, “Men and Women of the Corporation”, New York, Basic Books.
348 p.
33
Kaplan, S. and D. Reishus: 1990, “Outside Directorships and Corporate Performance”,
Journal of Financial Economics 27, 389-410.
Kesner, I. F.:1988, “Directors’ characteristics and committee membership: An
investigation of type, occupation, tenure and gender”, Academy of Management Journal
31, 66-84.
Klein, A.: 1998, “Firm Performance and Board Committee Structure”, Journal of Law
and Economics 41, 275-99.
Li, C. A. and B. Wearing: 2004, “Between glass ceilings: Female non-executive directors
in UK quoted companies”, International Journal of Disclosure and Governance 1 (4, Oct),
355-370.
Conference Board: 2005. “Emerging Trends in Europe… Women in Leadership. The
Diversity Imperative: Supporting Actions for Change”. By D. Palframan. Executive
Action Series No. 158 (August), 7 p.
Rotemberg, J. J. and G. Saloner: 1993, “Leadership Styles and Incentives,” Management
Science 39 (11), 1299-1318.
Shrader, C.B., V.B. Blackburn and P. Iles: 1997, “Women in Management and Firm
Financial Performance: An Exploratory Study”. Journal of Managerial Issues 9 (3) (Fall),
p. 355-372.
Tirole, J.: 2001, “Corporate Governance,” Econometrica 69 (1), pp. 1-35.
Westphal, J. D. and L. P. Milton: 2000, “How experience and network ties affect the
influence of demographic minorities on Corporate Board”. Administrative Science
Quarterly 45 (2), 366-417.
34