Does Board Gender Diversity Have a Financial Impact? Evidence
Using Stock Portfolio Performance
Larelle Chapple • Jacquelyn E. Humphrey
Received: 12 December 2012 / Accepted: 21 June 2013 / Published online: 9 July 2013
� Springer Science+Business Media Dordrecht 2013
Abstract There is growing regulatory pressure on firms
worldwide to address the under-representation of women in
senior positions. Regulators have taken a variety of approa-
ches to the issue. We investigate a jurisdiction that has issued
recommendations and disclosure requirements, rather than
implementing quotas. Much of the rhetoric surrounding gen-
der diversity centres on whether diversity has a financial
impact. In this paper we take an aggregate (market-level)
approach and compare the performance of portfolios of firms
with gender diverse boards to those without. We also inves-
tigate whether having multiple women on the board is linked
to performance, and if there is a within-industry effect.
Overall, we do not find evidence of an association between
diversity and performance. We find some weak evidence of a
negative correlation between having multiple women on the
board and performance, but that in some industries diversity is
positively correlated with performance.
Keywords Gender diversity � Corporate governance �
Financial performance
Introduction
Board gender diversity has become a widely debated cor-
porate governance topic over the last decade. Indeed, as
reported below, some capital market regulators have moved
to impose gender quotas on boards; whereas others have
taken a more ‘‘best practice’’ approach and provided rec-
ommendations and/or disclosure requirements on firms with
respect to their gender diversity profiles. We categorise these
approaches to corporate governance and board composition
as either ‘‘mandated’’ or ‘‘self-regulation’’. At the core of the
regulatory debate is the nature of the value to be created from
diverse boards and the practical pressure it imposes on firms
to respond to these corporate governance initiatives.
Individual firms may face the cost of changing their
board composition, or at least the cost of defending (dis-
closing) current practices. Academic literature from a
variety of disciplines argues that an initiative such as board
diversity is not solely an economic concern, but a matter
that also may appeal to various social factors, recognising
that firms participate not just in capital markets but in
society as a whole (Hafsi and Turgut 2013). However, the
intriguing proposition that motivates the narrow focus in
this study is the explicit statement from the Australian
market operator that improving gender diversity enhances
company performance—as an economic construct:
Research has shown that increased gender diversity
on boards is associated with better financial perfor-
mance, and that improved workforce participation at
all levels positively impacts on the economy. (ASX
20.
APM Welcome, APM North West Network Conference, Synergies Across Sectors
Does Board Gender Diversity Have a Financial Impact Evidence.docx
1. Does Board Gender Diversity Have a Financial Impact?
Evidence
Using Stock Portfolio Performance
Larelle Chapple • Jacquelyn E. Humphrey
Received: 12 December 2012 / Accepted: 21 June 2013 /
Published online: 9 July 2013
� Springer Science+Business Media Dordrecht 2013
Abstract There is growing regulatory pressure on firms
worldwide to address the under-representation of women in
senior positions. Regulators have taken a variety of approa-
ches to the issue. We investigate a jurisdiction that has issued
recommendations and disclosure requirements, rather than
implementing quotas. Much of the rhetoric surrounding gen-
der diversity centres on whether diversity has a financial
impact. In this paper we take an aggregate (market-level)
approach and compare the performance of portfolios of firms
with gender diverse boards to those without. We also inves-
tigate whether having multiple women on the board is linked
2. to performance, and if there is a within-industry effect.
Overall, we do not find evidence of an association between
diversity and performance. We find some weak evidence of a
negative correlation between having multiple women on the
board and performance, but that in some industries diversity is
positively correlated with performance.
Keywords Gender diversity � Corporate governance �
Financial performance
Introduction
Board gender diversity has become a widely debated cor-
porate governance topic over the last decade. Indeed, as
reported below, some capital market regulators have moved
to impose gender quotas on boards; whereas others have
taken a more ‘‘best practice’’ approach and provided rec-
ommendations and/or disclosure requirements on firms with
respect to their gender diversity profiles. We categorise these
approaches to corporate governance and board composition
as either ‘‘mandated’’ or ‘‘self-regulation’’. At the core of the
3. regulatory debate is the nature of the value to be created from
diverse boards and the practical pressure it imposes on firms
to respond to these corporate governance initiatives.
Individual firms may face the cost of changing their
board composition, or at least the cost of defending (dis-
closing) current practices. Academic literature from a
variety of disciplines argues that an initiative such as board
diversity is not solely an economic concern, but a matter
that also may appeal to various social factors, recognising
that firms participate not just in capital markets but in
society as a whole (Hafsi and Turgut 2013). However, the
intriguing proposition that motivates the narrow focus in
this study is the explicit statement from the Australian
market operator that improving gender diversity enhances
company performance—as an economic construct:
Research has shown that increased gender diversity
on boards is associated with better financial perfor-
mance, and that improved workforce participation at
4. all levels positively impacts on the economy. (ASX
2010)
1
Market operators, such as securities exchanges, rou-
tinely issue best practice corporate governance guidelines.
L. Chapple
QUT Business School, Queensland University of Technology,
Brisbane, QLD, Australia
J. E. Humphrey (&)
Research School of Finance, Actuarial Studies and Applied
Statistics, Australian National University, Level 4, College
of Business and Economics Building 26C, Canberra,
ACT 0200, Australia
e-mail: [email protected]
1
Australian Securities Exchange (2010), Corporate Governance
Principles and Recommendations with 2010 amendments.
Available
at
http://www.asxgroup.com.au/media/PDFs/cg_principles_recom
men
dations_with_2010_amendments.pdf. Accessed 16 November
5. 2012.
123
J Bus Ethics (2014) 122:709–723
DOI 10.1007/s10551-013-1785-0
http://www.asxgroup.com.au/media/PDFs/cg_principles_recom
mendations_with_2010_amendments.pdf
http://www.asxgroup.com.au/media/PDFs/cg_principles_recom
mendations_with_2010_amendments.pdf
In recent years, market operators and law makers have
begun to try to address the clear under-representation of
women in the upper echelons of the corporate world. Par-
ticularly, for example, in Australia and the UK, the market
operators recommend that listed companies disclose and
explain their chosen diversity policy and self-assessed
performance (ASX 2010; FRC 2012)
2
: a self-regulated
approach. An increased regulatory focus on board gender
diversity and improving the overall diversity of corporate
boards indicates that policy makers consider diversity at
6. the board level important. However, a natural question that
arises from the regulatory intervention is whether there is
any measurable economic result.
The aim of this study is to investigate the economic
impact of board gender diversity initiatives promulgated by
securities market operators, in a self-regulated environ-
ment. Prior literature on the economic impact of diversity
tends to focus on individual firm effects. This study extends
the literature by taking an overall or aggregate view of
financial performance in the capital markets. Specifically,
we construct portfolios of firms with gender diverse boards
and compare their returns over time to portfolios of firms
with all-male boards. Our innovation is that we take a
portfolio, or aggregate approach to measuring impact, not
just a firm-level approach.
Taking a portfolio approach also provides a substantial
econometric improvement over firm-level (panel) analysis.
Corporate governance research is plagued by endogeneity,
7. including issues caused by omitted variables, heterogeneity
among samples and reverse causality. Forming portfolios
means that firm-specific characteristics are averaged out,
eliminating both the heterogeneity issue and also reducing
the omitted variables problem—which arises because firm-
specific characteristics (independent variables) impact
firm-level outcomes (dependent variables). Further, pre-
sumably the market regulator is interested in the impact of
regulation on overall market outcomes, and not on specific
firms. Using portfolios also has the advantage of more
accurately reflecting how the new regulation will impact
the overall market on average, rather than the impact on
specific firms.
The Australian capital market setting is ideal for this
study as we are able to investigate the transition of a
market from purely voluntary board composition decisions,
through to more recent times in which the gender diversity
recommendation has been incorporated into best practice
8. guidelines. Although this became formal in 2010, there had
been considerable interest in and leakage of the proposed
regulation prior to this. The self-regulatory approach in
Australia is relevant to other market jurisdictions, for
example, the UK that has adopted a disclosure approach, as
well as those markets that remain unregulated on board
gender diversity, such as in North America. The alternative
approach is to mandate quotas, and this approach has been
taken in European countries such as France, Finland, Italy,
Spain, Norway and the Netherlands. There has been con-
siderable research interest in these markets as to the eco-
nomic benefits brought about by legislative mandate
(Torchia et al. 2011; Ahern and Dittmar 2012). The regu-
latory position can be a ‘‘moving target’’ of course: if the
persuasiveness of the guidelines does not affect real change
in Australian corporate boards, the Australian Discrimi-
nation Commissioner has signalled a move to mandatory
quotas in 2015 (AHRC 2010).
3
9. The descriptive statistics show the transition: in the
years of voluntary adoption of board diversity (between
2004 and 2010), the percentage of sample companies with
a diverse board hovered between 36 and 42 %. In the year
after self-regulation (2011), 52 % of the sample companies
reported a diverse board. One of the peak industry bodies,
the Australian Institute of Company Directors (AICD
2012), reports that it is only since 2010 that women have
been recruited to boards in relatively larger numbers.
4
This
move to self-regulation provides an ideal opportunity to
investigate questions of director diversity. Until the start of
this decade, there have been insufficient observations to
perform research using statistical techniques. For this rea-
son, we classify diversity as boards with one female
director, as opposed to none. Further discrimination is
difficult given the data, but nevertheless we believe this
10. analysis is extremely topical and timely.
In terms of our results, we find no compelling evidence
of a clear performance differential between firms with and
without female directors. However, we suggest that there
are plausible circumstances in which a firm that is larger,
more established and in a particular industry may ‘‘trade
up’’ to diversity as a business proposition, but not neces-
sarily for clear-cut quantifiable economic reasons.
Accordingly, we see the contribution of the research as
focussing on a particular aspect of the regulatory debate: to
measure the extent of the economic benefit, if any, to the
2
Ibid; Financial Reporting Council (2012), The UK Corporate
Governance Code. Available at
http://www.frc.org.uk/getattachment/
a7f0aa3a-57dd-4341-b3e8-ffa99899e154/UK-Corporate-
Governance-
Code-September-2012.aspx. Accessed 19 November 2012.
3
Australian Human Rights Commission (2010) Gender Equality
11. Blueprint 2010. Available at
http://www.humanrights.gov.au/sex_
discrimination/publication/blueprint/. Accessed 16 November
2012.
4
The AICD reports that in 2007 and 2008, the new appointments
to
boards who were female comprised 8% of appointments. In
2009, 5%,
2010, 25%, 2011, 28% and in 2012 it was 24%. The sample
population reported by the AICD comprises the S&P/ASX 200
(top
200 listed Australian companies). Our sample comprises the
S&P/
ASX 300 (see http://www.companydirectors.com.au/Director-
Resource-Centre/Governance-and-Director-Issues/Board-
Diversity/
Statistics. Accessed 14 March 2013).
710 L. Chapple, J. E. Humphrey
123
http://www.frc.org.uk/getattachment/a7f0aa3a-57dd-4341-b3e8-
ffa99899e154/UK-Corporate-Governance-Code-September-
2012.aspx
http://www.frc.org.uk/getattachment/a7f0aa3a-57dd-4341-b3e8-
13. odology in the fourth section. Results and analysis are
presented in the fifth section, and the final section con-
cludes with insights to the regulatory impetus.
Background and Literature
Regulatory Background
On 30 June 2010 the primary Australian market operator,
the Australian Securities Exchange, announced the latest
changes to its Corporate Governance principles (ASX
2010) recommending:
5
1. Companies should establish a policy concerning
diversity and disclose the policy or a summary of that
policy. The policy should include requirements for the
board to establish measurable objectives for achieving
gender diversity (and) for the board to assess annually
both the objectives and progress in achieving them.
2. Companies should disclose in each annual report the
measurable objectives for achieving gender diversity
set by the board in accordance with the diversity policy
14. and progress towards achieving them.
3. Companies should disclose in each annual report the
proportion of women employees in the whole organi-
sation, women in senior executive positions and
women on the board.
As an example, Table 1 demonstrates how a large
diversified firm listed on the ASX, Wesfarmers Ltd., has
complied with the these three aspects of recommended
disclosure. In its 2012 annual report, the company reported
its performance across measurable objectives (to foster an
inclusive culture; improve talent management; enhance
recruitment policies and ensure pay equity) and numerical
data on women employees, managers and directors.
6
The problem this corporate governance recommendation
seeks to address is the under-representation of women in
corporate management—particularly at the senior, board,
level. Australia is by no means alone in seeking to address
15. the lack of female representation; indeed, other countries
have implemented far more onerous requirements, for
example:
• France—women must hold 20 % of board positions by
2014, and 40 % by 2017.
• Italy—women must comprise 33 % of board positions
by 2015. Non-compliance will result in a fine of €1
million.
• Finland—companies have been required to have at least
one woman on the board since July 2010.
As previously noted, the ASX’s stated motivation for
implementing change to the corporate governance princi-
ples explicitly quotes enhanced company performance.
This claim by the ASX is puzzling because a positive
association between gender diversity and performance has
not been convincingly established in the available aca-
demic literature. The alternative explanations are that firms
will appoint women directors for other reasons—for
example, the firm is already performing well and there is
external pressure for diverse boards (Farrell and Hersch
16. 2005); or that ‘‘women may be being preferentially placed
in leadership roles that are associated with an increased risk
of negative consequence’’ (Ryan and Haslam 2005, p. 83).
Commentary from other disciplines argues a strong case
for moral or social legitimacy—firms choose gender
diverse boards for a variety of reasons and the ‘‘business
case’’ is not solely an economic imperative (Carter et al.
2010; Fairfax 2011), but offers a signal of the firm’s
commitment to its reputation and image to all stakeholders
(consumers, employees, etc.), not just investors (Burke
1997; Broome and Krawiec 2008). Below, we summarise
the academic literature on the impact of gender diversity on
firm performance and similar types of constructs.
Influence of Female Directors on Firm Performance
Most corporate governance research into gender diversity
that involves financial or market measures takes an agency
theory approach (Terjesen et al. 2009). According to agency
theory, the board of directors is the primary monitoring
17. mechanism to curb management’s tendency to behave in a
self-interested manner and not in the best interest of share-
holders (Hart 1995). Much research investigates the opti-
mum size, composition and characteristics of the board, but it
is accepted that firms with dispersed ownership (as typically
featured in market listed firms) should have boards that
comprise a majority of independent directors. Certainly,
despite any choices that firms may make on optimal board
composition (Hermalin and Weisbach 2003), many corpo-
rate governance codes now make a specific recommendation
on board independence composition.
5
Australian Securities Exchange (2010), op cit.
6
http://ir.wesfarmers.com.au/phoenix.zhtml?c=144042&p=irol-
reports
annual. Accessed 19 November 2012.
Does Board Gender Diversity Have a Financial Impact? 711
123
http://ir.wesfarmers.com.au/phoenix.zhtml?c=144042&p=irol-
18. reportsannual
http://ir.wesfarmers.com.au/phoenix.zhtml?c=144042&p=irol-
reportsannual
As to how women directors can positively influence firm
performance under the expectations of agency theory, for
example, whether female directors are ‘‘better’’ monitors,
is a matter of much conjecture in the literature. Using
agency theory as the perspective, gender should not matter
to board tasks (Nielsen and Huse 2010). The arguments can
at most be indirect. Adams and Ferreira (2009, p. 292)
suggest that ‘‘because they [female directors] do not belong
to the ‘old boys club’, female directors could more closely
correspond to the concept of the independent director
emphasized in theory.’’ Other studies focus on decision-
making tasks and suggest that female directors facilitate
communication in decision-making (Bilimoria 2000) or are
more risk averse in business decisions (Srinidhi et al.
2011). However, as Carter et al. (2003) point out, these
differences may also mitigate effective monitoring if the
19. ‘‘diverse’’ incumbents are marginalised on the board. As
the literature we rely on is ambivalent in its expressed prior
expectations, we do not hold a strong prior expectation.
Rather, the question as to gender diversity and financial
impact is motivated by the regulatory stance. More direct
prior evidence on gender diversity and financial perfor-
mance, predominantly based on firm-level analysis, is
presented below.
Relation Between Gender Diversity and Firm
Performance
Evidence of a direct association between a firm’s financial
performance and its board’s diversity profile remains elu-
sive. As Adams and Ferreira (2009, p. 305) assert: ‘‘The
literature on diversity also has ambiguous predictions for
the effect of diversity on performance.’’
A number of recent studies
7
that have investigated the
20. question empirically have not found consistent results.
Carter et al. (2010) do not find a significant relation
between firm performance (Tobin’s Q and ROA) and
diverse boards, using a sample of U.S. firms from the S&P
500 index for the period 1998–2002. Conversely, Carter
Table 1 An example of how one firm, Wesfarmers Limited, has
chosen to comply with the Australian regulator’s disclosure
requirements on
gender diversity
ASX
Principle
Format of disclosure
Diversity
policy
While Wesfarmers is committed to fostering all types of
diversity, gender diversity has and continues to be a priority for
the
Group. As set out in the Wesfarmers Diversity Policy, the
Group’s approach to gender diversity is based on four core
objectives:
foster an inclusive culture; improve talent management; enhance
recruitment practices; and ensure pay equity
21. Measurable
objectives
Foster an inclusive culture—Wesfarmers divisions undertake
different initiatives and practices based on the needs of their
business, such as flexible work practices at senior levels and
paid parental leave. Specific targets are linked to senior
executive
key performance objectives under the annual incentive plan
Improve talent management—at least once a year, the Group
Managing Director meets with each division to review: senior
leader
performance and development; succession plans for critical
roles; and the pipeline of high-potential leaders
During the 2012 financial year, talent reviews were conducted
with all divisions for senior manager level staff and above and
included 138 women.
This is in addition to detailed talent reviews conducted with
employees by individual businesses within the Wesfarmers
Group.
Throughout the Group, all high-potential leaders benefit from
an array of development opportunities such as internal and
external development programs, stretch assignments, action
learning projects, coaching, mentoring and 360� feedback
Enhance recruitment practices—in 2012, 37 % of externally
recruited positions and 30 % of internal promotions (all
22. manager
level and above roles) were filled by women
Ensure pay equity—a pay audit is conducted annually on a
Group basis (which includes a review of gender equity). Results
are
reviewed by the Board and divisional Managing Directors. In
addition, a pay equity review of all Wesfarmers divisions was
undertaken during the year, in line with previous years, which
did not indicate any observable discrepancies in pay across each
level, after taking into account performance, experience,
location and job nature
Proportion of women employees, management, directors
Percentage of female employees
30 June 2011 30 June 2012
Non-executive directors 25 25
Senior executive positions (general manager or above) 22 21
All management and professional roles 26 28
Total workforce 57 57
The table is extracted from Wesfarmers’ 2012 Annual Report
7
Although the studies are recent, the datasets are from the prior
decade.
23. 712 L. Chapple, J. E. Humphrey
123
et al. (2003), using a similar sample (Fortune 1000 firms in
1997), find a positive relation between performance (To-
bin’s Q) and diverse boards.
8
In a broader sample (S&P
1500, 1996–2003), Adams and Ferreira (2009) find some
evidence of a negative relation between gender diversity
and company performance, measured as both the ratio of
the firm’s market-to-book value (as a proxy for Tobin’s Q)
and ROA.
The studies to date using Australian data are highly con-
strained by sample size. Bonn (2004) found a positive rela-
tion between diversity and market-to-book ratio in a sample
drawn from top companies in 1999. In a sample of firms with
diverse boards in 2000–2001, diversity was associated with a
24. higher Tobin’s Q (Nguyen and Faff 2007). Wang and Clift
(2009) discuss an absence of any statistically significant
association between returns and the percentage of women on
boards, with diversity data for only 1 year (2003) for the top
500 listed companies, and using ROA, ROE and shareholder
return as performance measures.
9
We note that all these studies were undertaken prior to the
ASX’s diversity disclosure requirements and therefore have
extremely few firms with women directors. The change in
regulation and consequent increase in female director par-
ticipation justifies a re-examination of the Australian market.
Adams et al. (2011) examine director appointments and find
that the stock market reacts more favourably to the
appointment of women directors than men directors.
We take a slightly different approach from that used by
the current literature. Rather than examining diversity at
the firm level, we examine the aggregate returns generated
by portfolios of firms with diverse boards and compare
25. these both to non-diverse boards (all- male boards), as well
as boards with varying degrees of diversity (one woman
director on the board, or more than one woman director),
and within industry. Taking a higher level, aggregate,
approach is appropriate because we are particularly inter-
ested in the market regulator’s (high-level) perspective.
Essentially, we are investigating whether there is an asso-
ciation between gender diversity and overall market out-
comes, not just on specific firm outcomes.
Degrees of Gender Diversity
A gender diverse board has been simply defined as a board
with at least one female director (Adams and Ferreira 2009;
Campbell and Mınguez-Vera 2008), as this is the easiest
proxy to apply, especially given the lack of data (number of
firms with women directors) available. However, there is a
second-order research question: how diverse does a diverse
board have to be for there to be an economic impact? This has
been referred to as ‘‘critical mass’’ (Broome et al. 2011),
26. meaning that if there are enough women on a particular
board, women are no longer different from other board
members: the critical mass of the women directors no longer
makes them ‘‘outsiders’’. When applied to corporate boards,
prior research suggests that the critical mass for women
directors on a board is three or more (Konrad et al. 2008;
Torchia et al. 2011). Note, however, that the latter study was
conducted in Norway where there has been a female director
quota in place since 2005, hence enough variation in board
composition data are available to discriminate this measure.
In our study, due to the very low level of female board par-
ticipation in Australia, such variation in board composition is
not observable. We therefore choose to discriminate port-
folios as either no women directors, or at least one female
director (diversity); then, if there is diversity, whether there
is one female director or more than one female director.
Board Diversity and Other Firm-Level Economic
Enhancement
27. A ‘‘business case’’ for diversity can be approached by inves-
tigating other associations, typically at the firm level. Studies
in accounting examine financial outcomes such as earnings
quality (Krishnan and Parsons 2008), earnings management
(Srinidhi et al. 2011) and analyst forecast accuracy (Gul et al.
2011) as a function of diverse boards and/or management.
There are studies examining more direct economic impact
such as board diversity and cost of capital (Gul et al. 2009).
Finally, some research focuses on gender diversity effective-
ness on board performance, in the sense of board oversight and
monitoring (Hillman et al. 2008; Adams and Ferreira 2009).
This study examines a business case for the market
regulator—what does the capital market gain from
increased gender diversity on corporate boards? The aim of
our paper is to determine whether there is, in fact, any
association between financial performance and gender
diversity, by taking a market-level perspective. Specifically
we look at whether having women on a company’s board is
28. correlated with any performance advantage or disadvan-
tage. Of course, it could well be that the regulatory
imperative is not about economic performance but another
outcome—for example, social legitimacy (companies need
to conform to society’s rules, norms, etc.).
Data
Our initial sample comprises all firms listed on the S&P/
ASX 300 which broadly represents the largest 300 listed
8
Note both of these studies measure diversity as gender and
ethnicity.
9
The authors define ROE as the ratio of profit after interest and
tax to
book value of equity, whereas shareholder return is the ASX
realised
rate of return adjusted for dividends and splits.
Does Board Gender Diversity Have a Financial Impact? 713
123
29. Australian companies. Each month, we extract the con-
stituent list of S&P/ASX 300 firms from Datastream—this
allows us to track additions and deletions to the index over
our sample period. We extract returns, book-to-market and
market values of all stocks listed on the S&P/ASX 300
from Datastream.
We obtain data on companies’ board members from the
Boardroom database from Connect4. We verify the Con-
nect4 start and end dates of female board members by
checking company announcements on the ASX website.
10
We match our S&P/ASX 300 list of firms against the
available board member information from Connect4, which
gives us a final sample of 577 firms over our sample period.
As Connect4 data are only available from 2004 onwards,
our sample period is January 2004 to September 2011. We
use the return on the S&P/ASX 300 as a proxy for the
return on the market and extract these data from Data-
stream. The risk-free rate is proxied by the 90-day bank
30. accepted bill rate from the Reserve Bank of Australia.
11
Methodology
Portfolio Formation
As mentioned above, we investigate the economic impact of
gender diversity in a number of ways. We begin by splitting
our sample into two groups: firms with all-male boards and
firms with at least one woman on their board. We form port-
folios of these two groups, as represented in Fig. 1. We also
form a difference portfolio (the long/short portfolio), which is
long in the firms with women and short in the firms with only
men on the board. This is to allow us to clearly determine
whether there is a difference in the performance of firms with
gender diverse boards and those without. We form both value-
weighted and equally weighted portfolios. As already men-
tioned, we choose to take a portfolio approach because we are
interested in the market-level impact of gender diversity.
However, forming portfolios also has the added advantage of
31. diversifying firm-specific risk and improving the precision of
estimates from regression analysis.
We next investigate whether having more than one
woman is associated with differential performance. We
consequently split our female portfolio in two depending
upon whether there is one woman or more than one
woman on the board, also represented in Fig. 1.
12
We again
compare our portfolio of one woman (more than one
woman) to the portfolio of only men by forming a differ-
ence portfolio. We also compare whether there is a dif-
ference in the returns of the two female portfolios, i.e., we
create a difference portfolio between the one-woman and
more-than-one-woman portfolios.
Finally, prior research (for example, Brammer et al.
2007) has identified that women may be more value-rele-
vant in some industries than others.
13
32. We investigate this
proposition by forming portfolios comparing firms with
only men on the boards to firms that have at least one
female board member within industry using the ten
Industry Classification Benchmark (ICB) industry
classifications.
Empirical Framework
Our aim is to determine whether gender diversity is asso-
ciated with financial performance. We use a number of
widely accepted performance models to address this
question. First, we use a one-factor model as follows:
Rp;t � Rf;t ¼ ap þ bp Rm;t � Rf;t
� �
þ ep;t ð1Þ
where Rp;t , Rm;t and Rf;t are the returns on portfolio p, the
market portfolio and the risk-free asset in month t,
respectively.
A significantly positive (negative) alpha on our long/
short portfolios will indicate that gender diversity is cor-
33. related with an increase (decrease) in risk-adjusted
performance.
It is possible firms that choose to have women on their
boards may differ from those that do not in terms of their
size, book-to-market ratios and momentum exposures.
Specifically, larger firms may be more likely to hire women
as they may face external pressure to behave in a socially
acceptable way (Farrell and Hersch 2005), and have the
resources to do so. As these factors have been shown to
impact returns, it is important that we control for them
when investigating our portfolios. Consequently, we also
perform the analysis using a four-factor model:
Rp;t � Rf;t ¼ ap þ bpðRm;t � Rf;tÞþ spSMBt þ hpHMLt
þ upUMDt þ ep;t
ð2Þ
where Rp;t, Rm;t and Rf;t are as above and SMBt, HMLt and
UMDt are the monthly return on the mimicking size, book-
to-market and momentum factors.
34. We form size (SMB), book-to-market (HML) and
momentum (UMD) factors in line with Fama and French
(1993) and Carhart (1997). However, due to differences in
the Australian and U.S. tax systems, portfolios are formed
10
http://www.asx.com.au/asx/statistics/announcements.do.
Accessed
13 December 2012.
11
http://www.rba.gov.au/statistics/tables/index.html#interest_rate
s.
Accessed 12 June 2012.
12
Panel A of Table 2 clearly demonstrates that it is not possible to
disaggregate further on the number of women, since the vast
majority
of boards have only one or two women.
13
Note that Brammer et al. (2007) study is based on a UK sample.
714 L. Chapple, J. E. Humphrey
123
http://www.asx.com.au/asx/statistics/announcements.do
35. http://www.rba.gov.au/statistics/tables/index.html#interest_rate
s
in December of year t - 1, rather than in June, and held for
12 months (see also Gharghori et al. 2009). Each Decem-
ber, stocks are ranked on their market value and classified
as small or big depending upon whether their market value
is smaller or larger than the median market value. Inde-
pendently, stocks are ranked on their book-to-market ratio
and classified as either value, growth or neutral, with
breakpoints at the 30th and 70th percentile. Stocks with
negative book values are deleted. SMB is then calculated
as the value-weighted average return on the small portfo-
lios (small value, small neutral and small growth) minus
the value-weighted average return of the big portfolios (big
value, big neutral and big growth). Similarly, HML is the
average return of the high book-to-market portfolios (big
high and small high) minus the average return of the low
book-to-market portfolios (big low and small low). UMD is
36. formed in a similar way. At the end of December of year
t - 1, stocks are ranked on their prior 1-year return and
classified as up or down, using the 30th and 70th percentile
breakpoints. We use the same size classifications as before.
UMD is calculated as the average return of the up portfo-
lios (small up and big up) minus the average return of the
down portfolios (small down and big down).
Results and Analysis
As a useful snapshot of the gender diversity profile of the
boards of firms listed on the Australian capital market,
Table 2 presents descriptive statistics. Over the 8-year
duration of the sample representing the S&P/ASX 300,
there was an average of 287 firms (max 294; min 282).
Panel A shows that the percentage of the sample with at
least one female director fluctuates between 36 and 46 %
and peaks in the last year, 2011, at 52 %.
We are also interested in whether diversity can be dif-
ferentiated beyond a point of one female director, and the
37. data show that firms with two female directors fluctuate
around 10 % of the sample (except for 2011 where the
level is 17 %). The sample thins considerably beyond more
than two female directors.
In Panel B we report the raw returns on our initial
portfolios of all-male boards, boards with at least one
woman, and the market factors. The table shows that on an
equally weighted basis, the female portfolio underper-
forms, but the means of the value-weighted returns are
similar. However, neither of these differences is statisti-
cally significant: the p values on the paired t tests between
means (and Wilcoxon text between medians) are all above
0.30 for both the equally weighted and the value-weighted
portfolios.
Panel C shows the industry breakdown of the sample. It is
clear from the panel that boards with at least one female
director are concentrated in particular industries. The majority
of firms in consumer services, financials and telecommunica-
38. tions have at least one woman on their boards. However, firms
in basic materials are clearly dominated by all-male boards. It
must be noted that some industries, particularly telecommu-
nications, utilities, and technology comprise very few firms.
These small sample sizes need to be considered when we
interpret results from portfolios formed within these industries.
Table 3 presents regression results from portfolios of
firms with all-male boards compared to firms with at least
one female board member. We are mainly interested in the
long/short portfolio results because these indicate whether
there are significant differences between the portfolios. The
alphas on the long/short portfolios are insignificant,
meaning there is in fact no correlation between having
women on a firm’s board and returns. This result is upheld
regardless of whether we use the one- or four-factor model
or whether we value-weight or equally weight the
portfolios.
The long/short portfolios’ coefficients on the market
39. factor are significantly negative, which suggests that firms
that hire women tend to be of lower risk. We could perhaps
argue that perhaps the types of firms that hire women are
older, more established firms. In terms of loadings onto the
four factors, we see a significant negative loading on the
SMB factor for the long/short portfolios. This indicates it is
larger firms that tend to have women on their boards.
Perhaps this is to be expected: larger firms are more likely
to face external scrutiny and therefore feel pressurised to
take socially acceptable actions. Further, these firms are
likely to have the resources to do so. The HML factor is
significantly positive in the value-weighted portfolios. This
demonstrates that firms with women on boards tend to be
‘‘value’’ firms. We could argue that value firms are more
established firms that again may be expected to be able to
afford diversity. Coefficients on UMD are insignificant:
At least
one
woman
40. One
woman
More than
one
woman
At least
one
woman by
industry
All men
All men
by
industry
Fig. 1 This figure illustrates the way in which we divide our
sample
into the various portfolios. Each circle represents a portfolio
that is
composed of companies with the stated gender profile
Does Board Gender Diversity Have a Financial Impact? 715
123
having women on boards does not appear to be related to
41. prior stock performance.
14
Does Having Many Women Matter?
In Table 4 we divide our female portfolio into firms with
one woman versus those with more than one woman on the
board. In terms of differences between all-male boards and
boards with women, we again do not find any significant
alphas: there is no difference in the performance of boards
that have or do not have women. However, we are inter-
ested in whether there is a relationship between having one
or more than one woman on the board and returns. We find
weak evidence in our four-factor value-weighted results
that firms with more than one woman have lower returns
than firms with one woman on the board. However, we
note that this finding is only significant at the 10 % level
and is not consistent across all our tests. Our other long/
short portfolios do not display a significant difference in
the returns of firms with one versus more than one woman
on the board.
42. Firms with more than one woman on the board are larger
and tend to have more of a value tilt than firms with only
one woman on the board. This may again indicate that
firms that are more established are able to appoint more
women to their boards than other firms. The betas on the
two portfolios are not significantly different.
15
Does Industry Matter?
We investigate whether there is a correlation between
gender diverse boards and returns across different indus-
tries because prior research has identified that having
female board members is more valuable in some industries
than in others (see Brammer et al. 2007). Results are in
Table 5. For brevity, we present only results from the four-
factor model using value-weighted portfolios.
16
Focusing on the long/short portfolios, we see weakly
significantly positive alphas on basic materials and con-
sumer goods. It seems that in these two industries having at
43. least one woman on the board is associated with higher
returns.
17
However, for all eight other industries, we do not
find any correlation between having gender diverse boards
and returns: all the other alphas on the long/short portfolios
are insignificant.
The long/short portfolios’ coefficients on the size factor
are uniformly negative, albeit not always significant.
18
This
again highlights the fact that the larger firms appoint
women to their boards. Long/short portfolios’ loadings
onto the other factors (market, book-to-market and
momentum) are not as homogenous, with some positive
and some negative.
19
Caution needs to be exercised when interpreting some of
the industry results, however. A number of the industries
44. comprise very few firms, and dividing firms further into
those with and without women board members can dras-
tically reduce the number of firms in each portfolio. This is
particularly true of technology and telecommunications,
where there are some months with no observations in either
the female or the all-male portfolio. Results from basic
materials, financials, industrials and consumer services are
robust, however, as these portfolios comprise at least ten
firms in any given month.
20
Robustness Tests
In our main analysis, we form portfolios of firms with and
without women, as we believe this methodology best
demonstrates whether gender diversity provides higher
returns at the aggregate (market) level—as suggested by
the market regulator. This approach is perhaps unusual in
the corporate governance literature and therefore for
robustness we also perform the analysis in a panel set-
ting.
45. 21
In this case, we need firm-specific variables which
are only available on an annual basis.
We perform the analysis in two ways. First, we follow
Ahern and Dittmar (2012) and regress industry-adjusted
14
For robustness we also investigate the period prior to the ASX
recommendations. As discussions about this recommendation
were
already occurring in early 2009, we investigate the sample prior
to
2008. We rerun all regressions using the sample period January
2004
to December 2008. Results (not displayed, available upon
request) are
qualitatively identical. Alphas on the long/short portfolios are
insignificant and firms that have women on their boards are
larger,
value firms but do not load onto momentum.
15
We rerun the analysis using the period January 2004 to
December
46. 2008. Alphas on the long/short portfolios are uniformly
insignificant.
We do not find the size effect in the value-weighted long/short
portfolio but still find a significant value effect.
16
Other results available upon request.
17
The alphas on the long/short consumer goods portfolios are
significantly positive across all models although the alpha on
basic
materials is not significant in other specifications. The equally
weighted alphas on consumer services (telecommunications) are
significantly negative (positive) across the equally weighted
portfo-
lios. However, given that this result is not upheld in the value-
weighted models, this may be attributable to some poorly (over-
)
performing small firms in that sector.
18
Coefficients on the size factors are similarly predominantly
negative on the equally weighted portfolios.
19
47. Results for the January 2004 to December 2008 are similar with
most industries having insignificant alphas on the long/short
portfo-
lios. However, we do find outperformance in financials and
healthcare
and underperformance in industrials. Coefficients on SMB are
negative in the majority of the cases.
20
Both the all-male and the female consumer goods portfolios
have a
minimum of four firms in a particular month.
21
We thank an anonymous referee for this suggestion.
716 L. Chapple, J. E. Humphrey
123
Tobin’s Q against the proportion of women on the board as
the only independent variable, as well as time and period
fixed effects. We also investigate return on assets as the
dependent variable. Results (not displayed, available upon
request) show that in each case, the coefficient on the
48. proportion of women is insignificant.
We next perform an analysis similar to Adams and
Ferreira (2009) and regress the log of Tobin’s Q against:
proportion of women, board size, log revenue and the
proportion of non-executive directors. We follow those
authors and perform the regression in three ways: OLS,
firm fixed effects and then use an Arellano and Bond
Table 2 Descriptive statistics
Panel A
a
2004 2005 2006 2007 2008 2009 2010 2011
Number of firms 282 283 285 285 289 289 294 288
Firms with at least one woman 102 113 128 130 129 115 124
149
(percent of total) 36 % 40 % 45 % 46 % 45 % 40 % 42 % 52 %
Firms with
One woman 78 85 98 103 100 89 90 92
Two women 23 27 26 22 24 21 31 49
Three women 1 0 4 5 5 4 2 7
49. Four women 0 1 0 0 0 1 0 1
Five women 0 0 0 0 0 0 1 0
Panel B
b
Equally weighted Value weighted Market SMB HML UMD
Female All male Female All male
Mean 0.0056 0.0076 0.0068 0.0066 0.0029 -0.0003 0.0029
0.0005
Median 0.0152 0.0175 0.0150 0.0217 0.0138 -0.0045 0.0076 -
0.0004
Maximum 0.1234 0.2136 0.0769 0.0862 0.0733 0.0988 0.1405
0.1019
Minimum -0.2071 -0.2955 -0.0994 -0.2434 -0.1294 -0.1383 -
0.1089 -0.0829
SD 0.0499 0.0700 0.0372 0.0567 0.0406 0.0326 0.0394 0.0354
Panel C
c
Equally weighted Value weighted
All male Female All male Female
Basic materials 50.05 18.91 49.78 18.65
Consumer goods 11.09 6.52 11.24 6.48
50. Consumer services 10.07 23.01 10.10 22.96
Financials 24.26 35.41 24.57 35.39
Healthcare 9.62 6.24 9.68 6.23
Industrials 30.55 16.93 30.73 16.76
Oil & gas 15.92 6.62 16.02 6.46
Technology 5.39 4.01 5.48 4.01
Telecommunications 1.45 2.36 1.44 2.34
Utilities 4.61 4.00 4.67 3.96
a
This panel provides descriptive statistics on the firms in our
sample. Figures are as of 30 June in each year
b
This panel provides descriptive statistics on returns. Female
denotes portfolios of firms with at least one woman on the
board, All male are
portfolios of firms with no women on the board. Market, SMB,
HML and UMD denote the return on the market, size, book-to-
market and
momentum factor, respectively. Figures are per month. The
sample period is January 2004 to September 2011
c
This panel provides descriptive statistics on the industry
composition of the firms in our sample. Female denotes firms
51. that have at least one
female board member, All male denotes firms with no female
board members
Does Board Gender Diversity Have a Financial Impact? 717
123
(1991) dynamic panel model (i.e. including a lag of log
Tobin’s Q). Our results (not displayed, available upon
request) are similar to Adams and Ferreira (2009): the
coefficient on the proportion of women variable is insig-
nificantly positive using OLS, but significantly negative
using firm fixed effects and using a dynamic panel model.
As discussed in Adams and Ferreira (2009), these dispar-
ities in results highlight the necessity of correct model
specification that allows for potential endogeneity. We also
use return on assets as the dependent variable and in this
case the coefficient on the proportion of women is insig-
nificant in all specifications.
We conclude, then, that our robustness tests overall
52. uphold our main results: we do not find a relation between
having one or more women on the board and performance.
Discussion and Conclusion
The global regulatory interest in board gender diversity as a
corporate governance best practice guideline has escalated
over the last few years. Regulatory interest can manifest as
mandatory board quotas (for example, Norway), to recom-
mend and disclose regimes (for example, Australia and the
UK) to regulators who lag global practice (for example,
Canada and the U.S. which are as yet to address the issue).
This provides a range of environments for researchers to
study whether there is an association between diversity ini-
tiatives and firm value or outcomes. Most of the market-
based research referred to herein emanates from either the
mandatory environment (Norway) or the unregulated envi-
ronment (U.S.). There is literature from other disciplines that
Table 3 Returns on all firms
Alpha Market SMB HML UMD Adj R
53. 2
One-factor model
Equally weighted Female 0.003 1.15*** 0.87
(1.57) (20.08)
All male 0.006 1.53*** 0.78
(1.29) (12.06)
Long/short -0.003 -0.38*** 0.25
(-0.72) (-4.61)
Value weighted Female 0.004*** 0.90*** 0.98
(7.10) (53.50)
All male 0.004* 1.26*** 0.81
(1.73) (11.70)
Long/short 0.000 -0.35*** 0.20
(-0.14) (-2.89)
Four-factor model
Equally weighted Female 0.002* 1.01*** 0.36*** 0.23*** -
0.07* 0.94
(1.86) (37.58) (5.74) (6.77) (-1.82)
All male 0.005** 1.23*** 0.93*** 0.13** -0.08 0.95
54. (3.12) (22.94) (16.41) (2.27) (-1.41)
Long/short -0.003 -0.22*** -0.56*** 0.10 0.01 0.58
(-1.30) (-4.30) (-6.81) (1.18) (0.12)
Value weighted Female 0.004** 0.92*** -0.05** 0.01 0 0.98
(7.06) (56.52) (-2.03) (0.86) (-0.15)
All male 0.005** 1.16*** 0.37*** -0.23*** -0.06 0.88
(2.79) (17.85) (4.53) (-3.84) (-1.02)
Long/short -0.001 -0.24*** -0.42*** 0.24*** 0.06 0.47
(-0.48) (-3.04) (-4.21) (3.42) (0.82)
This table provides results for portfolios formed from our full
sample of firms. Female denotes portfolios of firms with at least
one female board
member, All male denotes portfolios of firms with no female
board members, Long/short is a portfolio long in Female and
short in Male. Alpha is
the alpha coefficient from the regression model. Market, SMB,
HML and UMD are the market, size, book-to-market and
momentum factors,
respectively. Newey-West HAC adjusted t statistics are in
parentheses. The sample period is January 2004 to September
2011. ***,**,* denote
significance at the 1, 5 and 10 % level, respectively
55. 718 L. Chapple, J. E. Humphrey
123
Table 4 Returns on firms with differing number of women
Alpha Market SMB HML UMD Adj R
2
One-factor model
Equally weighted One 0.003 1.19*** 0.86
(1.45) (16.62)
Many 0.0031 1.01*** 0.72
(1.25) (12.54)
One–many -0.0001 0.17 0.06
(-0.06) (1.57)
One–all male -0.0025 -0.34*** 0.23
(-0.76) (-4.82)
Many–all male -0.0024 -0.51*** 0.22
(-0.49) (-3.47)
Value weighted One 0.0053*** 0.98*** 0.91
57. Many–all male -0.003 -0.31*** -0.78*** 0.26*** -0.05 0.59
(-0.94) (-3.41) (-8.86) (2.92) (-0.61)
Value weighted One 0.0055*** 0.98*** 0.06 -0.09** 0.08* 0.92
(4.88) (26.36) (1.40) (-2.34) (1.72)
Many 0.0012 0.83*** -0.15*** 0.16** -0.09* 0.79
(0.70) (14.16) (-3.17) (2.04) (-1.79)
One–many 0.0044* 0.16* 0.21*** -0.26 0.17* 0.21
(1.70) (1.76) (2.88) (-2.19) (1.90)
One–all male 0.0008 -0.17*** -0.31*** 0.13*** 0.14* 0.36
(0.41) (-2.72) (-3.11) (2.81) (1.79)
Many–all male -0.0035 -0.33*** -0.52*** 0.39*** -0.03 0.44
(-1.18) (-2.83) (-4.56) (2.95) -(0.35)
This table provides results for portfolios formed from our full
sample of firms. One denotes portfolios of firms with one
female board member,
Many denotes portfolios of firms with more than one female
board member, All male denotes portfolios of firms with no
female board members.
Alpha is the alpha coefficient from the regression model.
Market, SMB, HML and UMD are the market, size, book-to-
market and momentum
58. factors, respectively. Newey-West HAC adjusted t statistics are
in parentheses. The sample period is January 2004 to September
2011. ***,**,*
denote significance at the 1, 5 and 10 % level, respectively
Does Board Gender Diversity Have a Financial Impact? 719
123
Table 5 Returns on firms within industries
Alpha Market SMB HML UMD Adj R
2
Basic materials Female 0.0097*** 1.09*** 0.07 -0.49***
0.25** 0.68
(3.32) (12.79) (0.47) (-4.29) (2.49)
All Male 0.0025 0.93*** 0.37*** -0.05 0.02 0.75
(0.99) (10.36) (2.75) (-0.74) (0.23)
Long/short 0.0072* 0.17 -0.30 -0.45*** 0.24* 0.15
(1.73) (1.21) (-1.30) (-3.04) (1.78)
Consumer goods Female 0.0412*** 3.31*** 0.15 -0.79 0.61
0.51
(2.85) (8.11) (0.27) (-1.57) (1.26)
61. (1.34) (2.89) (-1.52) (0.84) (-3.12)
All Male 0.1375 10.81** 2.16 2.12 -0.96 0.13
(0.93) (2.08) (0.66) (0.74) (-0.34)
Long/short -0.0168 -1.45 -7.65 -0.24 -7.54* -0.01
(-0.08) (-0.21) (-1.56) (-0.06) (-1.77)
720 L. Chapple, J. E. Humphrey
123
uses other evidence, such as surveys and interviews, to
examine the firm-level impacts of diverse boards.
This study is set in the Australian market because it has
recently introduced a ‘‘soft’’ regulatory approach—a rec-
ommendation that listed firms establish a gender diversity
policy and disclose their performance and achievements
against their adopted policy. Hence, the environment is not
mandatory, but creates strong external pressure to conform.
Second, rather perplexingly, the Australian market operator
has motivated its stance by claiming the ‘‘business case’’—
62. that diversity is linked to performance. However, despite
several studies investigating this problem, a conclusive link
between firm performance and gender diversity has not
been established.
We use a different methodology from prior literature
and take a portfolio approach to the question. This aggre-
gate approach more appropriately reflects the high-level
view relevant to the regulator. One shortcoming of our
study is one common to many studies reported herein—the
extremely small proportion of women on Australian boards
and, consequently, the low number of firms with female
board members available. We have been able to ameliorate
this constraint to some extent by using a portfolio
approach. However, to investigate the degree of diversity,
we are only able to discriminate between diversity meaning
one female director, compared to diversity being more than
one female director. The descriptive data show that the
percentage of boards with two female directors is around
63. 10 % for most of our sample period, but beyond that point
(three or more women) the sample thins considerably. This
constraint may weaken our ability to detect whether gender
diversity truly has a financial impact (it could be that larger
numbers of female directors on boards are necessary for
real ‘‘change’’), but perhaps demonstrates why the market
regulator has chosen to intervene in this area.
There are several industry reports that narrate the raw
data on board composition in the Australian capital market.
We are able to confirm that the percentage of diverse
boards (at least one female director) in the top 300 firms
fluctuates around 36–46 % and peaks in 2011 (the first year
after the corporate governance regulation) at 52 %. As
expected, there are industry clusters, but the industry
clusters may be hard to predict.
22
We find the majority of
diverse boards in the top 300 firms are in the consumer
services, financials and telecommunications industries.
64. Overall, we do not find a strong business case for gender
diversity on boards. Regardless of whether we use a one- or
four-factor model, we find no difference in the performance
of gender diverse and all-male board portfolios. However,
we find weak evidence (significant in one of our models
only) that more than one woman on a board is associated
with lower returns. The absence of strong return results is
informative to the market operator; it confirms that it is
difficult to find evidence of an economic argument for
diversity using market data.
However, the study also contributes in terms of empir-
ical evidence to support other value-relevant propositions
Table 5 continued
Alpha Market SMB HML UMD Adj R
2
Telecommunications Female -0.0047*** 0.01** 0.01* 0.01* 0
0.14
(-26.41) (2.15) (1.73) (1.75) (0.88)
All Male 0.0611 4.22*** 1.63 -2.41* 1.29 0.18
65. (1.20) (3.24) (1.15) (-1.99) (1.54)
Long/short -0.0657 -4.22*** -1.63 2.42* -1.29 0.18
(-1.29) (-3.23) (-1.15) (1.99) (-1.53)
Utilities Female 0.0015 0.8** 0.14 -0.47 0.07 0.07
(0.15) (2.44) (0.21) (-1.27) (0.15)
All Male 0.0021 0.36*** 0.26* 0.03 -0.11 0.20
(0.47) (3.58) (1.75) (0.28) (-1.13)
Long/short -0.0006 0.44 -0.12 -0.50 0.18 -0.00
(-0.05) (1.18) (-0.17) (-1.24) (0.38)
This table provides results for value-weighted portfolios formed
within each of ten industries. Female denotes portfolios of firms
with at least one
female board member, All male denotes portfolios of firms with
no female board members, Long/short is a portfolio long in
Female and short in
All Male. Alpha is the alpha coefficient from the regression
model. Market, SMB, HML and UMD the market, size, book-to-
market and
momentum factors, respectively. Newey-West HAC adjusted t
statistics are in parentheses. The sample period is January 2004
to September
2011. ***,**,* denote significance at the 1, 5 and 10 % level,
66. respectively
22
Adams et al. (2011) use government data from the Equal
Opportunity for Women in the Workplace Agency to predict that
high participation rates in the workforce may affect diversity,
so that
finance has a high workplace participation, whereas the natural
resources sector is male-dominated.
Does Board Gender Diversity Have a Financial Impact? 721
123
about diverse boards: we find larger firms with lower risk
tend to have diverse boards, and it is the very large firms
that are able to have more than one female director. This
suggests that firms that are established can ‘‘afford’’ diverse
boards. Further, there is weak evidence that having at least
one female director is correlated with higher returns for the
basic materials and consumer goods industries.
Our findings suggest that in a non-mandated environ-
67. ment, evidence of a link between diverse boards and
financial returns is elusive. A diverse board may be one
corporate governance mechanism that a firm can ‘‘trade
up’’ to for a range of complex societal reasons or stake-
holder expectations, but neither firms nor the regulator
should expect diversity to be associated with increased
stock returns. Of course, this study has cited a selection of
studies from a variety of disciplines and it may well be that
gender diversity is not solely a ‘‘business case’’ but a ‘‘buy-
in’’ for a range of reasons. The nature of these alternative
motivations for gender diverse boards is an interesting
empirical question that we leave to future research. Given
the potential suite of non-financial motives, we would
predict that the absence of market evidence would not
preclude a move to a more mandated approach to the
appointment of women to corporate boards.
Acknowledgments The authors thank John Nowland, Emma
Schultz, Tom Smith, Garry Twite and workshop participants at
the
68. Australian National University for helpful comments. We also
thank
Chen Cheng and Theingi Oo for research assistance. We thank
Susan
McCreery for proofreading the manuscript. We acknowledge the
Accounting and Finance Association of Australia and New
Zealand
for financial support.
References
Adams, R., & Ferreira, D. (2009). Women in the boardroom and
their
impact on governance and performance. Journal of Financial
Economics, 94(2), 291–309.
Adams, R., Gray, S., & Nowland, J. (2011). Does gender matter
in the
boardroom? Evidence from the market reaction to mandatory
new director announcements. Working Paper, University of
Queensland.
Ahern, K., & Dittmar, A. (2012). The changing of the boards:
The
impact on firm valuation of mandated female board representa-
69. tion. Quarterly Journal of Economics, 127(1), 137–197.
Arellano, M., & Bond, S. (1991). Some tests of specification for
panel
data: Monte Carlo evidence and an application to employment
equations. Review of Economic Studies, 58(2), 277–297.
Australian Institute of Company Directors. (2012).
Appointments to
ASX200 Boards.
Bilimoria, D. (2000). Building the business case for women
corporate
directors. In R. J. Burke & M. C. Mattis (Eds.), Women on
corporate boards of directors: International challenges and
opportunities (pp. 25–40). Dordrecht: Kluwer Academic
Publishers.
Bonn, I. (2004). Board structure and firm performance:
Evidence
from Australia. Journal of the Australian and New Zealand
Academy of Management, 10(1), 14–24.
Brammer, S., Millington, A., & Pavelin, S. (2007). Gender and
ethnic
diversity among UK corporate boards. Corporate Governance:
70. An International Review, 15(2), 393–403.
Broome, L. L., Conley, J. M., & Krawiec, K. D. (2011). Does
critical
mass matter? Views from the boardroom. Working Paper,
University of North Carolina.
Broome, L. L., & Krawiec, K. D. (2008). Signaling through
board
diversity: Is anyone listening? University of Cincinnati Law
Review, 77(1), 446–447.
Burke, R. (1997). Women on corporate boards of directors: A
needed
resource. Journal of Business Ethics, 16(9), 909–915.
Campbell, K., & Mınguez-Vera, A. (2008). Gender diversity in
the
boardroom and firm financial performance. Journal of Business
Ethics, 83(3), 435–451.
Carhart, M. M. (1997). On persistence in mutual fund
performance.
Journal of Finance, 52(1), 57–82.
Carter, D., D’Souza, F., Simkins, B., & Simpson, W. (2010).
The
71. gender and ethnic diversity of US boards and board committees
and firm financial performance. Corporate Governance: An
International Review, 18(5), 396–414.
Carter, D., Simkins, B., & Simpson, W. G. (2003). Corporate
governance, board diversity, and firm value. Financial Review,
38(1), 33–53.
Fairfax, L. (2011). Board diversity revisited: New rationale,
same old
story? North Carolina Law Review, 89(1), 855–886.
Fama, E. F., & French, K. R. (1993). Common risk factors in
the
returns on stocks and bonds. Journal of Financial Economics,
33(1), 3–56.
Farrell, K. A., & Hersch, P. L. (2005). Additions to corporate
boards:
The effect of gender. Journal of Corporate Finance, 11(1–2),
85–106.
Gharghori, P., Lee, R., & Veeraraghavan, M. (2009). Anomalies
and
stock returns: Australian evidence. Accounting and Finance,
72. 49(3), 555–576.
Gul, F. A., Chung-ki, M., & Srinidhi, B. (2009). Gender
diversity on
U.S. corporate boards and cost of capital. Working Paper, Hong
Kong Polytechnic University.
Gul, F. A., Hutchinson, M., & Lai, K. (2011). Gender diversity
and
properties of analyst earnings forecasts. Working Paper,
Monash
University.
Hafsi, T., & Turgut, G. (2013). Boardroom diversity and its
effect on
social performance: Conceptualization and empirical evidence.
Journal of Business Ethics, 112(2), 463–479.
Hart, O. (1995). Corporate governance: Some theory and
implica-
tions. The Economic Journal, 105, 678–689.
Hermalin, B. E., & Weisbach, M. S. (2003). Boards of directors
as an
endogenously determined institution: A survey of the economic
literature. FRBNY Economic Policy Review, 9(1), 7–26.
Hillman, A. J., Nicholson, G. J., & Shropshire, C. (2008).
73. Directors’
multiple identities, identification, and board monitoring and
resource provision. Organization Science, 19(3), 441–456.
Konrad, A. M., Kramer, V. W., & Erkut, S. (2008). Critical
mass: The
impact of three or more women on corporate boards. Organi-
zational Dynamics, 37(2), 145–164.
Krishnan, G. V., & Parsons, L. M. (2008). Getting to the bottom
line:
An exploration of gender and earnings quality. Journal of
Business Ethics, 78(1–2), 65–76.
Nguyen, H., & Faff, R. (2007). Impact of board size and board
diversity on firm value: Australian evidence. Corporate Owner-
ship and Control, 4(2), 24–32.
Nielsen, S., & Huse, M. (2010). The contribution of women on
boards
of directors: Going beyond the surface. Corporate Governance:
An International Review, 18(2), 136–148.
722 L. Chapple, J. E. Humphrey
123
74. Ryan, M., & Haslam, S. A. (2005). The glass cliff: Evidence
that
women are over-represented in precarious leadership positions.
British Journal of Management, 16(2), 81–90.
Srinidhi, B., Gul, F. A., & Tsui, J. (2011). Female directors and
earnings quality. Contemporary Accounting Research, 28(5),
1610–1644.
Terjesen, S., Sealy, R., & Singh, V. (2009). Women directors
and
corporate boards: A review and research agenda. Corporate
Governance: An International Review, 17(3), 320–337.
Torchia, M., Calabro, A., & Huse, M. (2011). Women directors
on
corporate boards: From tokenism to critical mass. Journal of
Business Ethics, 102(2), 299–317.
Wang, Y., & Clift, B. (2009). Is there a ‘‘business case’’ for
board
diversity. Pacific Accounting Review, 21(2), 88–103.
Does Board Gender Diversity Have a Financial Impact? 723
75. 123
Copyright of Journal of Business Ethics is the property of
Springer Science & Business Media
B.V. and its content may not be copied or emailed to multiple
sites or posted to a listserv
without the copyright holder's express written permission.
However, users may print,
download, or email articles for individual use.
Does Board Gender Diversity Have a Financial Impact?
Evidence Using Stock Portfolio
PerformanceAbstractIntroductionBackground and
LiteratureRegulatory BackgroundInfluence of Female Directors
on Firm PerformanceRelation Between Gender Diversity and
Firm PerformanceDegrees of Gender DiversityBoard Diversity
and Other Firm-Level Economic
EnhancementDataMethodologyPortfolio FormationEmpirical
FrameworkResults and AnalysisDoes Having Many Women
Matter?Does Industry Matter?Robustness TestsDiscussion and
ConclusionAcknowledgmentsReferences
The impact of racial and gender diversity in management on
financial
performance: how participative strategy making features can
unleash
a diversity advantage
Orlando C. Richard
a
*, Susan L. Kirby
76. b
and Ken Chadwick
c
a
Jindal School of Management, Department of Organizations,
Strategy, and International
Management, University of Texas at Dallas, Richardson, TX,
USA;
b
McCoy College of Business
Administration, Texas State University, San Marcos, TX, USA;
c
Department of Management,
Marketing, and Business Administration, Nichols State
University, Thibodaux, LA, USA
How does racial and gender diversity in the management ranks
affect the bottom line?
Our findings indicate that participative strategy making (PSM)
positively moderates the
relationship between both racial and gender diversity in
management and firm
performance measured as return on assets. Specifically, PSM
strengthens the positive
relationship that exists between racial diversity in management
and firm performance.
Although no main effect is observed for gender diversity in
management, our results
reveal that gender diversity in management is positively related
to performance when
77. PSM is high. However, we find that gender diversity in
management is negatively
related to performance when PSM is low, while gender
homogeneous management
experience superior performance. We offer implications for
diversity research to
embrace and consider the role of PSM and ‘inclusiveness’.
Keywords: gender diversity in management; inclusiveness;
knowledge-based view;
participative management; racial diversity; racial diversity in
management
Introduction
Theories predicting and research examining the impact of
diversity in organizations
continues to be challenged by inconclusive results (e.g.
Harrison and Klein 2007; van
Knippenberg and Schippers 2007; Joshi and Roh 2009).
According to Harrison and Klein
(2007), although research on the effects of diversity in
organizations has increased
dramatically in recent years, consistent findings remain elusive.
Similarly, Joshi and Roh
(2009), in a meta-analysis examining the impact of contextual
factors in research on team
diversity, state that a majority of studies report a
nonsignificant, direct relationship between
78. diversity and performance with the effects of gender, race, age
and tenure diversity on
performance varying. In response to these mixed results,
researchers have sought to refine
theories by examining moderators of the relationship between
diversity and performance
(Wegge, Roth, Kanfer, Neubach and Schmidt 2008; Joshi and
Roh 2009).
In addition, race and sex, among the most visible and salient of
cultural diversity
dimensions, have been underdeveloped in studies exploring
strategic decision-making within
organizations(Nkomo1992;Richard2000;RobersonandKulik2007
)althoughitisprojected
that women and racial minorities will increasingly enter the
managerial ranks (U.S.
Department of Labor, Bureau of Labor Statistics 2009). This
shift, in combination with the
q 2013 Taylor & Francis
*Corresponding author. Email: [email protected]
The International Journal of Human Resource Management,
2013
Vol. 24, No. 13, 2571–2582,
http://dx.doi.org/10.1080/09585192.2012.744335
79. increasing use of management teams throughout organizations
(Kirchmeyer and McLellan
1991) and the inconsistent results, suggests the need to further
examine the economic impact
ofmanagementdiversity,aswellastheembeddedstrategymakingfeat
uresthatmayprovidea
firm with capabilities to leverage their diversity for competitive
advantage.
Previous research reveals that demographic compositions are
associated with increased
creativity and a broader foundation on which to base decisions
(Bantel and Jackson 1989);
thereby having tangible positive effects on firm outcomes
(Knight et al. 1999). Hambrick
and Mason (1984) suggest that organizational strategic
outcomes and processes are indeed a
function of characteristics of an organization’s managers.
However, research has also found
that management heterogeneity affects performance when
managers are given discretion or
latitude of action in making decisions as well as have low job
demands placed upon them
(Finkelstein and Hambrick 1990; Hambrick, Finkelstein and
80. Mooney 2005; Hambrick
2007). Therefore, we believe participative strategy making
(PSM) is essential for the
facilitation of positive interactions among racially and gender
diverse managers.
Theoretical background and hypotheses
Diversity can be broadly defined as a construct representing the
distribution of differences
among members of a unit with respect to a specific attribute
(Jackson, Joshi and Erhardt
2003; Harrison and Klein 2007; van Knippenberg and Schippers
2007). Readily
detectable, or relations-oriented, differences include attributes
such as race, gender and
age (Joshi and Roh 2009). We focus on these rarely explored
dimensions of diversity to
contribute to the existing body of diversity research and more
importantly to go beyond
studies investigating team performance by exploring bottom-
line performance effects.
Racial and gender diversity in management as knowledge-based
resources
The main tenet of the knowledge-based view is that expansive
expertise, based on a broad
81. range of individual experiences, is the key to organizational
functioning. Therefore,
organizational financial performance depends on leveraging
expertise to achieve competitive
advantage via know-how that competitors cannot easily imitate
or obtain (Barney 1991).
Miller and Shamsie (1996) categorize such resources as the
creative or collaborative skills
that enhance a firm’s ability to both develop and market
competitive products.
Racial and gender heterogeneity results in wider ranging
perspectives, diverse types of
information and ideas within an organization and results in
superior problem solving and
decision-making (Bantel and Jackson 1989; Cox, Lobel and
McLeod 1991; Pelled,
Eisenhardt and Xin 1999; Richard, McMillan, Chadwick and
Dwyer 2003). Individuals
from diverse backgrounds bring the organization multiple
perspectives for problem
solving and strategy formulation (Williams and O’Reilly 1998;
Jehn, Northcraft and Neale
1999). For instance, we propose that women and racial
minorities offer ‘requisite variety’
82. in organizations (Ashby 1956) and this variety leads to
increased communication and
performance benefits that are the result of creativity and
improved decision-making
(Zenger and Lawrence 1989; Watson, Kumar and Michaelsen
1993; Wiersema and Bantel
1993; Milliken and Martins 1996). For example, McLeod and
Lobel (1992) found that
groups with heterogeneous ethnic backgrounds produced higher
quality ideas and
generated a greater range of perspectives and alternatives in
brainstorming tasks than did
more homogeneous groups. Also, Cox et al. (1991) found that
ethnically diverse groups
made choices reflecting different cultural orientations in a two-
party prisoner’s dilemma
O.C. Richard et al.2572
game (Earley 1989). These ethnic and cultural perspectives may
provide organizations
with unique insights into distinct local niches as well as
potential international markets.
In sum, diversity can promote creativity and improves decision-
83. making, and hence,
lead to superior performance (Pelled et al. 1999; Richard and
Shelor 2002; Richard,
Barnett, Dwyer and Chadwick 2004). In the appropriate context,
managerial racial and
gender diversity can result not only in sociocultural benefits
(Cohen and Garcia 2008), but
also financial returns (Dwyer, Richard and Chadwick 2003). We
offer PSM processes as a
crucial process needed to forcefully unleash the positive
benefits of managerial diversity.
The role of PSM
Scholars emphasize that for firms to benefit from diversity, they
must emphasize
inclusiveness within the organization (Richard, Kochan and
McMillan-Capehart 2002;
Dwyer et al. 2003; Pless and Maak 2004). For example, Pless
and Maak (2004) emphasize
how an organizational culture of inclusion, which allows people
with different
backgrounds, mindsets and ways of thinking to work together, is
critical to unleashing
any potential ‘diversity advantage’. In addition, Swann, Polzer,
Seyle and Ko (2004)
84. describe how a group atmosphere that encourages freedom of
expression facilitates
openness among diverse members. Also, Shore et al. (2011)
emphasize that practices,
which promote sharing information, participation in decision-
making and voice, should be
reflected in measures of inclusiveness. We present PSM as a
measure of inclusiveness that,
we argue, facilitates positive interaction among diverse
management members. We focus
on these features as opposed to other more wide scale features
(e.g. diversity climates and
organizational cultures) because PSM is a construct that was
designed and validated
specifically for studying the vertical and horizontal strategy
making processes within the
management corps (Dess, Lumpkin and Covin 1997).
Diversity can be a double-edged sword, increasing the
opportunity for creativity as well
as the likelihood of communication difficulties and
misunderstandings or faultlines (Jehn
1995). According to the social identity perspective, racial and
gender heterogeneity hinders
85. group cohesiveness (Williams and O’Reilly 1998; Carson,
Mosley and Boyar 2004).
However, the present research posits that differences within
groups can be bridged. In order
for diverse groups to be effective and efficient, they must be
able to avoid destructive
faultlines (Lau and Murnighan 1998) so they can reach
consensus regarding group decisions.
Participative processes serve as one mechanism that may allow
organizations to
leverage their creative variety emanating from diversity while
preventing the negative
consequences of potential faultlines. Research by van
Knippenberg, De Dreu and Homans
(2004) emphasizes that in order for diversity to be advantageous
there must be a
participative process in place that allows for multiple
viewpoints to be considered. PSM
processes not only foster collaboration and socialization across
levels of management but
also between or among diverse members. PSM should therefore
facilitate knowledge
sharing across diverse management groups, an important feature
needed to truly benefit
86. from diversity and obtain a ‘diversity advantage’ (De Carolis
2003). Increased
communication and collaboration among organizational
members is the key to success
of organizations by allowing firms to utilize divergent
technical, creative or collaborative
skills and improve decision-making effectiveness through the
pooling and integration of
group resources. Therefore, the pitfalls associated with diversity
can be diminished and the
benefits can be amplified when PSM processes exist.
Given the potential benefits and costs of workplace diversity
(Milliken and Martins
1996), and the potential for participative strategic making
processes to improve
The International Journal of Human Resource Management 2573
organizational financial performance, we predict that group
heterogeneity alone may not
be advantageous if a firm is unable to take advantage of the
unique insight, judgment,
experience and know-how of women and minorities (Shrader,
Blackburn and Iles 1997).
87. Participatory strategy formulation processes are one way firms
can leverage workforce
diversity and cultivate creativity, insight and capabilities.
Participatory strategy
formulation incorporates broad ranges of perspectives,
knowledge, values and skills,
giving groups better tools for effective decision-making even in
complex environments
(Carmelli, Sheaffer and Halevi 2009). Therefore, we predict:
Hypothesis 1a: PSM positively moderates the racial diversity in
management to firm
performance relationship such that racial diversity is more
positively
related to performance when PSM is high rather than low.
Hypothesis 1b: PSM positively moderates the gender diversity
in management to firm
performance relationship such that gender diversity is more
positively
related to performance when PSM is high rather than low.
Methodology
Sample
We utilized a stratified random sample frame consisting of 700
banking institutions with
88. $500 million or more in total assets, 700 with $100–$499
million in total assets and 700
with $100 million or less in total assets (Dwyer et al. 2003;
Richard et al. 2004). We
mailed survey instruments to 2100 bank presidents and received
response from 535. The
chief executive officer completed items related to the firms’
PSM process, the analysis
strategic orientation and their personal demographic
background. We then obtained
contact information for the senior human resource (HR)
executives of the 535 banks that
responded to the initial survey and queried these HR executives
regarding the racial and
gender composition of each bank’s management, the firms’ PSM
process and personal
demographic attributes. Of the 535 HR executives surveyed, 168
provided the required
data (31% response rate). A logistic regression revealed that
respondents did not differ
from those who did not respond on asset size and return on
assets (ROA). On average,
participating banks had 154 employees (31 managers), 7
branches and were 77 years old.
89. Measures
Independent variables
To assess the racial and gender diversity of each bank’s
management composition, we
provided each HR executive a blank Equal Employment
Opportunity-1 Standard Form
100, which allowed us to collect demographic data. Multiple
layers of management are
involved in strategic making process (Birkinshaw 1997) so we
used the management job
category that includes administrative and managerial personnel
who set broad policies,
exercise overall responsibility for the execution of these
policies and direct individual
departments or special phases of a firm’s operations (Equal
Employment Opportunity
Commission [EEOC] 2010).
Blau’s (1977) index of heterogeneity was used to develop
measures of racial (white,
black, Asian, Hispanic and Native American) and gender (men
and women) diversity in
management consistent with previous research (Richard et al.
2004). Blau’s index
90. (calculated as 1 2
P
Pi
2
, where P is the proportion of individuals in a category and i
(1 2 n) is the number of categories) could thus theoretically
range from 0 to 0.80. In our
sample, Blau’s index values for race ranged from 0 to 0.61. For
managerial gender
O.C. Richard et al.2574
heterogeneity, we observed index values from 0 to 0.50,
reflecting the entire theoretical
range.
To measure PSM process, we utilized the five-point Likert-type
scale ranging from (1)
strongly disagree to (5) strongly agree (Dess et al. 1997). The
CEO and VP of HRM
responses on the PSM scale converge ( p , 0.001) so we
averaged the two responses for
each bank to create a more valid measure. The coefficient alpha
for the 10-item scale was
0.75. Sample items are ‘Cooperation and collaboration across
91. functional roles are actively
encouraged’, ‘People with unpopular views are given a fair
hearing in this bank’,
‘Working in this bank is like being part of a team’ and
‘Decisions concerning business
strategy are made on a consensus basis, involving people from
many departments’.
Control variables
Firm size was operationalized as the logarithmic transformation
of total assets. We also
included a dummy to control for bank holding company
affiliation (1 ¼ holding company;
0 ¼ other) consistent with Delery and Doty (1996). In addition,
geographic scope could
covary with managerial diversity, so it was captured as number
of branches divided by
total assets (Richard 2000). Bank age, measured as number of
years in services, was also
included as a control variable. Consistent with previous
demography research (Pelled et al.
1999), we also control for group size measured as number of
officials and managers.
Racial heterogeneity and the proportion of whites in
management are not synonymous. For
92. example, two groups, one with 10% Asian Americans and 90%
African American and
another with 10% Asian American and 90% White, would have
the same diversity score,
but would not be the same in terms of ‘whiteness’. Although our
sample closely resembles
that in the banking industry in that the majority of management
is likely to be White,
including a proportional control variable allows us to
disentangle results of our racial
diversity index more appropriately (Richard et al. 2004).
Therefore, we controlled for the
percentage of Whites in management. We also controlled for the
percentage of men in
management.
Dependent variable
ROA represented our measure of performance. It was year-end
while the demographic
data was year-beginning.
Results
Table 1 shows the means, standard deviations and correlations
for all measures. To test our
93. moderating hypotheses, we follow an analytical procedure
employed by Dess et al. (1997).
Hierarchical regression results for the test of the hypotheses are
shown in Table 2.
Hypotheses 1a and 1b state that PSM positively moderates the
managerial diversity to
financial performance relationship. Model 1 in Table 2 reports
the results for the control
variables, while Model 2 includes controls along with race and
gender diversity in
management measures. Model 3 inputs the hypothesized PSM
moderator and Model 4
adds the two-way interaction term to test H1a and H1b. For H1a
related to racial
heterogeneity in management, there was a positive interaction
effect on ROA ( p , 0.1
Model 4; p , 0.01 Race Alone Model). For gender diversity in
management (H1b),
support was also found on our ROA measure ( p , 0.01 Model 4;
p , 0.001 Gender Alone
Model). Therefore, both H1a and H1b are supported. For clarity,
we developed graphs of
the moderating effects.
The International Journal of Human Resource Management 2575
146. a
il
e
d
).
The International Journal of Human Resource Management 2577
Figure 1 reflects the positive performance effects from the
coupling of racial diversity
in management with PSM. Although low levels of PSM show a
slight positive effect for
racial diversity, the strongest effect emanates at high levels of
PSM. Figure 2 illustrates the
interaction effect of gender diversity in management and PSM
on the ROA measure,
which reveals the highest level of ROA at high levels of gender
diversity along with high
levels of PSM. Interestingly, gender homogeneous management
perform superior to
diverse ones at low PSM levels.
Discussion
Social identity theorist would propose that diversity results in
conflict but participative
settings, which can promote superordinate goals can help
147. diminish this conflict.
Information and decision-making perspectives would suggest
diversity has positive
benefits only when properly managed so we set out to explore
the role of participative
management in unleashing a diversity advantage. Both together
suggest that participative
management structures might simultaneously reduce conflict
and increase knowledge
sharing. The present research provides empirical evidence that
racial and gender diversity
in management effects on firm performance are leveraged by
inclusiveness (Dwyer et al.
2003; Shore et al. 2011), specifically PSM processes. Racially
homogeneous management
did not benefit from high levels of PSM and gender
homogeneous groups actually
experienced weaker firm performance. We conclude that if a
firm has diversity along with
full participation of all participants’ (regarding racial or gender
majority or minority
status) knowledge, skills and abilities (i.e. inclusiveness), they
will experience superior
performance. Interestingly, our study is one of the first to show
148. that the isolated interaction
affects between gender diversity in management and PSM were
stronger (7% increase in R
2
)
Figure 1. Two-way interaction between racial diversity in
management and participative strategy
making on return on assets.
O.C. Richard et al.2578
than race effects (5% increase in R
2
). Thus, PSM must be even more critical to have in the
managementrankswhenthereisgenderdiversity.
As this study reveals, linear models do not hold much promise
(e.g. marginal main
effects for racial diversity) and that the consideration of
contextual moderators offer the
most empirical support (Joshi and Roh 2009). Future research
should explore more
complex ideal types and actual bundles of types of diversity,
strategy making processes,
HR management (HRM) practices and environmental factors
(Doty, Glick and Huber
149. 1993; Delery and Doty 1996). In other words, more extensive,
high-order, configuration
designs warrant consideration in the organizational demography
literature. In addition,
although our findings have implications for organization
competitiveness, especially in the
U.S. financial services industry whose recent actions (i.e.
irresponsible lending practices,
aggressive and risky monetary policies) contributed to a
financial crisis and almost
collapse of the U.S. economy (Wheelock 2010), future research
should explore the impact
of diversity in an international context and across industries.
Finally, our results show that
PSM features can make a difference (e.g. Kreiser and Davis
2010; Short, Broberg,
Cogliser and Brigham 2010) and appears to be a key feature
needed to reap positive
diversity effects.
We offer several limitations of the current research. First, we
were not able to assess
diversity separately within each of the three levels of
management (i.e. upper, middle and
lower). There could be additional variance explained across
150. each of these levels that we
are unable to capture in our design. Future research should
address this issue.
Notwithstanding, we believe the present way we operationalize
management is
appropriate since the goal was to explore levels of PSM and
knowledge sharing across
levels of management (i.e. vertical integration) and not just
within a particular level (i.e.
horizontal integration), which ultimately allows us to more
rigorously test our framework.
Indeed, both PSM was designed for the level of analysis in
which we tested it. Second, the
Figure 2. Two-way interaction between gender diversity in
management and participative strategy
making on return on assets.
The International Journal of Human Resource Management 2579
cross-sectional nature of the data negates the ability to assess
the long-term effects of
diversity and more confidently employ a causal model. Recent
research suggests that racial
diversity may have different long-term effects on measures of
performance rather than
151. more intermediate measures (Richard, Murthi and Ismail 2007).
In addition, the sample
was limited in scope to U.S. banks. Consequently, any
generalization of results to other
settings should be done so with caution.
Organizations seeking to understand how to effectively manage
workforce diversity,
particularly racial and gender diversity are often faced with how
to avoid experiencing
negative effects such as interpersonal conflict, reduced
cohesion, distrust and decreased
task performance that evolve from social identity theory
explanations (Williams and
O’Reilly 1998). Current logic suggests that if negative effects
emanating from
demographic differences can be diminished, the beneficial
processes that allow knowledge
sharing among diverse organizational members can be exploited
for competitive advantage
(van Knippenberg et al. 2004; Harrison and Klein 2007). Our
research acknowledges
one essential strategy making process that allows diverse
management groups to fully
152. contribute, thereby improving firm performance, but other
process should be explored
(Hart and Banbury 1994).
First, organizations need to recognize that managerial diversity
represents a valuable,
knowledge-based HR that can contribute to competitive
advantage. Second, organizations
should design a PSM process in which cooperation and
collaboration across functional
roles are actively encouraged and unpopular views are given a
fair hearing. Put differently,
firms must create an inclusive process that promotes
belongingness (Shore et al. 2011). By
accounting for the participative processes that encourage
inclusiveness in diverse groups,
we have shed light on how organizations can avoid potential
diversity pitfalls and obtain a
superior diversity advantage.
References
Ashby, W.R. (1956), An Introduction to Cybernetics, London:
Chapman & Hall.
Bantel, K.A., and Jackson, S.E. (1989), ‘Top Management and
Annovation in Banking: Does the
Composition of the Top Team Make a Difference?’ Strategic
153. Management Journal, 10,
107–124.
Barney, J.B. (1991), ‘Firm Resources and Sustained
Competitive Advantage,’ Journal of
Management, 17, 99–120.
Birkinshaw, J. (1997), ‘Entrepreneurship in Multinational
Corporations: The Characteristics of
Subsidiary Initiatives,’ Strategic Management Journal, 18, 207–
229.
Blau, P.M. (1977), Heterogeneity and Inequality: A Primitive
Theory of Social Structure, New York:
Free Press.
Carmelli, A., Sheaffer, Z., and Halevi, M.Y. (2009), ‘Does
Participatory Decision-Making in Top
Management Teams Enhance Decision Effectiveness and Firm
Performance,’ Personnel
Review, 38, 696–714.
Carson, C.M., Mosley, D.C., and Boyar, S.L. (2004),
‘Performance Gains Through Diverse Top
Management Teams,’ Team Performance Management, 10, 121–
126.
Cohen, G.L., and Garcia, J. (2008), ‘Identity, Belonging, and
Achievement: A Model, Interventions,
Implications,’ Current Directions in Psychological Science, 17,
365–369.
Cox, T.H., Lobel, S.A., and McLeod, P. (1991), ‘Effects of
Ethnic Group Cultural Differences on
Cooperative and Competitive Behavior on a Group Task,’
Academy of Management Journal, 34,
154. 827–847.
De Carolis, D.M. (2003), ‘Competencies and Imitability in the
Pharmaceutical Industry: An Analysis
of Their Relationship With Firm Performance,’ Journal of
Management, 29, 27–50.
Delery, J.E., and Doty, D.H. (1996), ‘Modes of Theorizing in
Strategic Human Resource
Management: Tests of Universalistic, Contingency, and
Configurational Performance
Predictions,’ Academy of Management Journal, 39, 802–835.
O.C. Richard et al.2580
Dess, G.G., Lumpkin, G.T., and Covin, J.G. (1997),
‘Entrepreneurial Strategy Making and Firm
Performance: Tests of Contingency and Configurational
Models,’ Strategic Management
Journal, 18, 677–695.
Doty, D.H., Glick, W.H., and Huber, G.F. (1993), ‘Fit,
Equifinality, and Organizational
Effectiveness: A Test of Two Configurational Theories,’
Academy of Management Journal, 36,
1196–1250.
Dwyer, S., Richard, O.C., and Chadwick, K. (2003), ‘Gender
Diversity in Management and Firm
Performance: The Influence of Growth Orientation and
Organizational Culture,’ Journal of
Business Research, 56, 1009–1019.
Earley, P.C. (1989), ‘Social Loafing and Collectivism,’