Article Review HRM

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

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_recommen

dations_with_2010_amendments.pdf. Accessed 16 November 2012.

123

J Bus Ethics (2014) 122:709–723

DOI 10.1007/s10551-013-1785-0

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

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,

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

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

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

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

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

market of the regulatory initiative. That the results do not

show superior portfolio returns for diverse boards is still

informative to the regulatory debate and the academic

challenge to identify and measure the benefits and

enhancements brought about by board diversity.

This paper proceeds as follows. The background and

literature is reviewed and contextualised in the next section

and data described in section three. We present the meth-

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

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

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

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

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

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

‘‘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

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

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 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.

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

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

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),

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

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

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

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

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

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

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-

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.

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_rates.

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

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

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

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-

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

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

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

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.

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

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

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. 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

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

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

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

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

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 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

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 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

(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

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

(4.35) (24.83)

Many 0.0016 0.81*** 0.75

(0.85) (10.75)

One–many 0.0037 0.17 0.04

(1.26) (1.50)

One–all male 0.0012 -0.28*** 0.17

(0.52) (-3.23)

Many–all male -0.0025 -0.45** 0.17

(-0.62) (-2.51)

Four-factor model

Equally weighted One 0.0024* 1.04*** 0.43*** 0.17*** -0.05 0.94

(1.85) (27.86) (5.96) (4.11) (-1.29)

Many 0.002 0.92*** 0.14** 0.39*** -0.13* 0.82

(0.83) (17.26) (2.22) (7.45) (-1.83)

One–many 0.0004 0.12 0.29*** -0.22*** 0.08 0.30

(0.17) (1.61) (4.07) (-3.21) (1.17)

One–all male -0.0026 -0.19*** -0.49*** 0.04 0.03 0.52

(-1.22) (-4.14) (-5.64) (0.49) (0.47)

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

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)

All Male 0.0064 1.31*** 0.54* 0.33 0.07 0.35

(0.75) (4.62) (1.75) (1.08) (0.37)

Long/short 0.0347* 2.01*** -0.39 -1.12* 0.54 0.19

(1.93) (4.7) (-0.51) (-1.88) (0.98)

Consumer services Female -0.0034** 0.52*** 0.07* 0.07** -0.06 0.73

(-2.49) (11.88) (1.67) (2.01) (-1.08)

All Male -0.0056*** 0.48*** 0.39*** 0.25*** -0.02 0.67

(-3.32) (5.58) (2.96) (4.14) (-0.34)

Long/short 0.0022 0.04 -0.32** -0.19** -0.05 0.16

(0.96) (0.50) (-2.23) (-2.51) (-0.58)

Financials Female 0.0019 0.69*** -0.08 0.14*** -0.07 0.77

(1.59) (12.87) (-1.45) (2.76) (-1.55)

All Male -0.0018 2.3*** 0.88*** -0.65 -0.79*** 0.66

(-0.23) (5.50) (3.46) (-1.37) (-4.68)

Long/short 0.0037 -1.61*** -0.96*** 0.80 0.72*** 0.52

(0.46) (-3.61) (-3.39) (1.54) (3.85)

Healthcare Female 0.0156 4.45*** -0.8 1.78*** -0.16 0.73

(1.17) (9.76) (-1.38) (4.10) (-0.39)

All Male 0.0155 1.90*** 0.25 -1.96*** -0.86* 0.44

(1.34) (5.63) (0.97) (-3.25) (-1.81)

Long/short 0.0001 2.55*** -1.05* 3.74*** 0.71 0.47

(0.01) (3.93) (-1.75) (4.03) (0.96)

Industrials Female 0.0286*** 3.16*** -0.67** 1.21*** -0.04 0.62

(2.76) (10.52) (-2.31) (3.47) (-0.11)

All Male 0.0087* 0.68*** -0.05 -0.07 0.10 0.25

(1.94) (5.23) (-0.41) (-0.65) (0.88)

Long/short 0.0199 2.48*** -0.62* 1.28*** -0.14 0.48

(1.62) (7.06) (-1.85) (3.58) (-0.34)

Oil & gas Female -0.0023 0.45*** 0.15 0.01 -0.10* 0.46

(-0.9) (5.78) (1.38) (0.14) (-1.81)

All Male 0.0059 0.58*** 0.28 -0.30 0.10 0.17

(1.05) (3.16) (1.43) (-1.57) (0.66)

Long/short -0.0082 -0.13 -0.13 0.31 -0.21 0.01

(-1.3) (-0.61) (-0.52) (1.37) (-1.33)

Technology Female 0.1545 9.65*** -5.66 2.11 -8.85*** 0.12

(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’’—

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

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.

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

(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, 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-

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

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-

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:

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

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,

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). 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

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

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.