board diversity and financial performance in the top 500 ... · board diversity and financial...

Board Diversity and Financial Performance in the Top 500 Australian Firms

Alireza Vafaei, Department of Accounting, La Trobe Business School, La Trobe University

Kamran Ahmed, Department of Accounting, La Trobe Business School, La Trobe University

Paul Mather, Department of Accounting, La Trobe Business School, La Trobe University

Recent regulatory changes in Australia require listed companies to implement policies for increasing board diversity

and to report thereon. In this paper we examine the association between gender diversity on corporate boards and the

financial performance of a large sample of the top 500 listed companies in Australia during the period 2005–2011,

addressing many of the methodological weaknesses in prior studies. Our descriptive statistics show that the proportion

of female directors on boards increased markedly between 2010, when ASX amended principles came into effect, and

2011. The results also show that board diversity is positively associated with financial performance after controlling

for a number of firm-specific, ownership and governance characteristics and potential endogeneity with the two-stage

least square tests. Thus, we inform the policy debate by providing empirical evidence supporting the business case for

board diversity.

The board of directors is the key decision-making

body in a corporation and is responsible for approving important strategic operational and financial decisions (Ferreira 2010). In most countries including Australia, women hold relatively few highly visible decision-making positions (Rubin 2000; Jurkus et al. 2011), although female board membership is in-creasing (Adams and Ferreira 2009). The current debate questions whether or not governments should implement regulations to increase female participation in senior management (e.g., Dale-Olsen et al. 2013). We in-form this debate by examining the association between gender diversity on corporate boards (here after ‘board diversity’) and the financial performance of Australian listed companies.

The advocates for increasing board diversity make ar-

guments that fall into two broad categories: ‘fairness and

equity’ and ‘shareholder value and firm performance’.

Studies that investigate the appointment of women on

boards from a procedural justice perspective (e.g., Mc-

Cann and Wheeler 2011; Brook and Tyler 2011) suggest

that people care about the fairness of decision making

above and beyond the outcomes derived from those de-

cisions (Brook and Tyler 2011). A procedural justice ar-

gument justifies female appointments on the basis that

gender equity is a basic human right. It is about the place

of women in society and the role of the corporate sec-tor

in recognising and supporting them (McCann and

Wheeler 2011). The studies that investigate the appoint-

ment of women on boards from a shareholder value and

firm performance or business case perspective (e.g.,

Erhardt et al. 2003; Carter et al. 2003; Nguyen and Faff

2007; Gul et al. 2013) argue that board diversity

improves organisational value and performance by

providing the board with new vision and perspectives.

Gul et al. (2013: 512) indicate that board diversity leads

to a greater knowledge base, creativity and innovation

and therefore provides higher competitive advantage to

the firm. Consistent with the current drive for gender balance

within corporations in many developed and emerging

countries, the Australian Securities Exchange (ASX)

Corporate Governance Council has made a number of

amendments to its principles and recommendations

(hereafter ‘ASX amended principles’) requiring com-

panies to establish policies for, and increase reporting on,

board diversity (ASX Corporate Governance Council

2010).1 Similarly, the Australian Institute of Company

Directors announced a range of measures encouraging the

appointment of women directors (Australian Institute of

Company Directors 2009) but these have received mixed

reactions, with the structure and capacity of smaller ASX

listed firms to diversify their boards being questioned

(Korporaal 2010). Empirical studies investigating the association

between board diversity and firm financial

performance are inconclusive. For example, some Correspondence: Professor Kamran Ahmed, Department of

Accounting, La Trobe Business School, Faculty of Business,

Economics and Law, La Trobe University, Victoria 3086,

Australia. Tel: +61 3 9479 1125; fax: +61 3 9479 2356; email:

[email protected]

Australian Accounting Review No. 75 Vol. 25 Issue 4 2015 doi: 10.1111/auar.12068 413

Board Diversity and Financial Performance

empirical studies report that a higher proportion of

women in top management have a statistically

significant positive effect on firm performance (Carter

et al. 2003; Erhardt et al. 2003; Bonn 2004; Krishnan

and Park 2005; Nguyen and Faff 2007; Campbell and

Minquez-Vera 2008) while others register a negative

effect (e.g., Adams and Ferreira 2009; Ahren and

Dittmar 2012). Other studies have found no influence on performance by gender diversity in

top management (Shrader et al. 1997; Moncrief et al. 2000; Mohan and Chen 2004; Smith et al. 2006;

Rose 2007; Wang and Clift 2009; Eklund et al. 2009).

The inconclusive findings of prior research,

especially the three Australian studies (Bonn 2004;

Nguyen and Faff 2007; Wang and Clift 2009) may be

due to a number of methodological problems. These

issues include failure to test for reverse causality

(endogeneity), un-observed firm heterogeneity, not

controlling for governance factors affecting firm

performance, relatively small samples and use of a

narrow proxy for performance measures. In this paper, we address the main limitations in

previous Australian studies by utilising a large sample of

the top 500 ASX listed firms over a more recent and

longer time period (2005–2011) by employing multiple

measures of firm performance and board diversity, and by

addressing the endogeneity issue in a systematic way. We

find that board diversity has improved particularly since

2010 in Australian listed firms and there is a positive and

significant association between female non-executives on

boards and firm financial performance. Our study is

timely and contributes to the understanding of the role of

female directors in firm performance by providing

empirical evidence on the expected economic efficacy of

the ASX amended principles. It informs an important

policy debate and should be of interest to regulators,

practitioners and academics. The paper is structured as follows. Section 2 provides a

literature review and hypothesis development. The re-

search methodology, variable measurement and regres-

sion models are presented in Section 3. Descriptive statis-

tics are presented in Section 4, followed by a presentation

of the results and further analysis in Section 5. Section 6

concludes the paper.

Literature Review and Hypothesis Development

Rationales in favour of female appointments are

dominated by business case arguments (see, e.g., Cox and

Blake 1991; Robinson and Dechant 1997; Carter et al.

2003; Gul et al. 2011; Jurkus et al. 2011; Joecks et al.

2013; Gul et al. 2013) with social and procedural justice

arguments (see, e.g., Brooke and Tyler 2011), being less

A. Vafaei, K. Ahmed & P. Mather

prominent (McCann and Wheeler 2011). Carter et al.

(2003) find a significant and positive association be-

tween the percentage of women on the board of directors

and firm performance using a sample of 638 publicly

traded Fortune 1000 firms. Erhardt et al. (2003) find a

positive association between demographic diversity on

boards and accounting measures of performance. Miller

and Triana (2009) find that gender-heterogeneous boards

identify new innovative opportunities and produce higher

quality decisions compared to those that are

homogeneous. Gul et al. (2011), using stock price in-

formativeness as an alternative measure of performance,

report a positive association between gender diversity on

corporate boards and firm performance. Their analysis

suggests that gender diversity improves stock price

informativeness by increasing voluntary public

disclosures in large firms and increasing the incentives

for private information collection in small firms. Other studies (including Smith et al. 2006; Campbel and

Minguez-Vera 2008; Dunstan et al. 2011; Mahadeo et al.

2012; Luckerath-Rovers 2013) also indicate a significant and

positive association between gender diversity and firm

performance. Adams and Ferreira (2009: 291) find that

female directors have better attendance records than male

directors and gender-diverse boards allocate more effort to

monitoring. However, they report on average that gender

diversity has a negative effect on firm performance, which

is, according to them, driven by companies with fewer

takeover defences. Jurkus et al. (2011), using agency

arguments, suggest that female directors are shown as

‘tough’ monitors and find that firms with a greater presence

of female officers have lower agency costs. However, the

negative relation is not robust when considering the

endogeneity of diversity suggesting that endogeneity issues

need to be addressed to derive reliable results. Earlier,

Farrell and Hersch (2005), in a sample of large Fortune 500

firms, failed to find convincing evidence that board gender

diversity is a value enhancing strategy. This is supported by

Lee and James (2007), who report that market reactions to

the announcements of female chief executive officers

(CEOs) are more negative than those of their male coun-

terparts. Haslam et al. (2010) find that women are found on

the boards of companies that are perceived to be performing

poorly and that their presence on boards can lead to the

devaluation of companies by investors in the United

Kingdom. Recent studies such as that by Carter et al. (2010),

using three-stage least squares (3SLS) regression analysis,

indicate no significant association between gender and ethnic

diversity and firm performance, measured by Tobin’s Q and

ROA. Joecks et al. (2013) suggest that the association

between performance and board diversity is not linear as

previously suggested in the literature. They find that gender

diversity initially affects firm performance negatively but

after reaching a critical mass of about 30% firm performance

improves.

414 Australian Accounting Review

C 2015 CPA Australia


In Australia the results are similarly inconclusive. For

example, Bonn (2004) finds a positive relationship be-

tween the proportion of female directors in 1999 and

market-to-book ratio in a sample of 84 large ASX listed

firms. Later, Nguyen and Faff (2007) investigated the as-

sociation between board diversity (measured by both

dummy and continuous variables) and firm performance

(measured by Tobin’s Q) for a sample of 832 firm-year

observations over the period 2000–2001. Their results

indicate that gender-diverse boards are rewarded with a

value premium as reflected in higher market valuation.

Wang and Clift (2009) indicate no significant association

between gender and ethnic diversity on the board and

financial performance. There are several theoretical arguments (e.g., Pfeffer

1973; Bantel and Jackson 1989; Murray 1989; Carter et

al. 2003) that support board diversity on corporate boards

and assert that board diversity would lead to better firm

performance. For example, Pfeffer (1973), employing

resource dependence theory, argues that incorporation of

diverse constituencies and stakeholders into the board

facilitates the acquisition of critical re-sources for the

organisation. Hillman et al. (2000) based on Pfeffer

(1973) provide a list of benefits with some commonly

observed characteristics of directors and suggest that

different types of directors would provide various

valuable resources to the firm. As a result, a more diverse

board would provide more beneficial resources, which

should produce better firm performance (Carter et al.

2010). Agency theory has also been used to argue that

diversity increases board independence because people

with a different gender, ethnicity or cultural background

might ask questions that would not come from directors

with more traditional backgrounds (Carter et al. 2003).

Later, Jurkus et al. (2011) argue that board diversity

reduces agency costs which, in turn, improve firm

performance, especially for those firms with a weak

governance structure. Overall, drawing on the theoretical arguments

derived from resource dependence and agency theory

as well as the relative weight of prior international

literature dis-cussed earlier, we expect that board

diversity will impact positively on firm financial

performance. Accordingly, we hypothesise that:

H1: The presence of women on corporate boards is

positively associated with firm financial performance. Research Methodology, Variable Measurement and Regression Models

Sample

The initial sample for this study comprised the top 500

ASX listed firms ranked by market capitalisation over

the period 2005–2011 (3500 firm-year observations).


Table 1 Sample selection procedure and sample composition by industry Panel A: Sample selection Top 500 ASX listed companies (2005–2010) 3500

Less missing financial data (586)

Less missing corporate governance data (1813)

Total 1101

Panel B: Sample composition by industry

Materials 25.4%

Financials 22%

Industrials 14.2%

Consumer discretionary 11.2%

Energy 10.6%

Consumer staples 4.6%

Health care 4.6%

Utilities 3.8%

Information technology 2%

Telecommunications 1.6%

At 31 December 2003, the top 500 listed companies

represented 95% of the total market capitalisation of the

ASX listed firms (Standard and Poor’s 2004; Wang and

Clift 2009). Therefore, this dataset covers a very large

proportion of the total ASX market capitalisation. Owing

to missing financial data, 586 firm-year observations

were excluded, reducing the sample size to 2914 firm-

year observations. A further 1813 firm-year observations

with missing corporate governance data, including board

diversity, board size and the number of non-executive

directors, decreased the sample size to a final unbalanced

panel of 1101 firm-year observations. Panel A of Table 1

summarises the sample selection procedure. The sources

of data include the Connect 4 Boardroom Review

database for the characteristics of the directors of the top

500 ASX listed firms, the Connect 4 database containing

the annual reports of the top 500 ASX listed firms,

DatAnalysis, which provides company data, market

information and statistics of ASX listed firms, and

Securities Industry Research Centre of Asia Pacific

(SIRCA). The latter is used for collection of two control

variables: block-holder ownership (BHO) and managerial

share ownership. As indicated in Panel B of Table 1, the

highest proportion of sampled firms are from the

materials sector, 25.4%, followed by financials and

industrials at 22% and 14.2% respectively.

Variable measurement

We employ four measures of firm performance to increase the reliability of the results. These are: two accounting-based measures (ROA and ROE), a market-based measure (Tobin’s Q) and an economic-based

measure of firm performance (CFO/TA).2 Accounting-

based measures of firm performance are based on an assessment of how the company has performed in the

C 2015 CPA Australia Australian Accounting Review 415


past, while market-based measures indicate the current

position of a company and its potential in the future

(Wang and Clift 2009; Haslam et al. 2010). ROA is

measured by dividing operating earnings by total assets

and ROE is measured by dividing net profit after tax

before abnormals by ordinary shareholders’ equity.

Tobin’s Q is measured as the sum of market value of

equity and book value of liabilities divided by the book

value of total assets at the balance sheet date. This

simple version of Tobin’s Q is applied widely in

Australia (Nguyen and Faff 2007) and elsewhere

(Haslam et al. 2010).3 Finally, following Healy et al.

(1992) and Givoly and Hayn (2000) CFO/TA is

measured by dividing cash flow from operations

(CFO) by total assets at the balance sheet date. Following Jurkus et al. (2011) female membership on

boards is measured by a dummy variable (Div) indicating

their presence and a continuous variable (Divprop),

reflecting the percentage of female directors on the board.

Campbell and Minguez-Vera (2008) and Wang and Clift

(2009) suggest that the mixed evidence on the association

between board composition and firm performance could

be attributed to the omission of other variables that affect

performance. As a result, and consistent with a sub-set of

the extant literature that uses a relatively comprehensive

set of control variables (see, e.g., Ander-son and Reeb

2003; Nguyen and Faff 2007; Adams and Ferreira 2009),

a range of governance and other variables are used as

controls. These are block-holder ownership, managerial

share-ownership, board size, firm age, lever-age, firm

size, proportion of non-executive directors and assets-in-

place. Block-holder ownership (BHO) is defined as share-

holders who own more than 5% of a company’s common

shares (Morck et al. 1988). Managerial share-ownership

(MSO) is the percentage of shares owned by directors at

the reporting date (Short and Keasey 1999). Board size

(BSize) is measured as the natural logarithm of the total

number of directors on the board (Goodstein et al. 1994;

Dalton et al. 1999). Firm age (FAge) is the number of

years the firm has been listed on the ASX (Majumder

1997; Baker and Kennedy 2002). Leverage (Lev) is

measured by dividing the book value of long-term debt by

the market value of equity (Wruck 1990; Henry 2008).

Firm size (FSize) is measured as the natural logarithm of

total assets (Gooding and Wagner III 1985; Smith et al.

2006). The proportion of non-executive directors

(NonExec) is measured by dividing the number of non-

executive di-rectors by the total number of directors on

the board (Pearce and Zahra 1992; Hart 1995; Agrawal

and Knoeber 1996). Finally, assets-in-place is used as a

proxy for the investment opportunity set and is used as an

inverse growth option indicator. Following Zingales

(2000) and Frye (2004), we measure this variable as the

ratio of the sum of inventory, property, plant and

equipment to total assets.


Regression models

To investigate the association between board diversity and firm performance, we use two measures of board

diversity Divit and Divpropit, and four measures of

firm performance, that is, ROAit, ROEit, Tobin’s Qit

and CFO/TAit. The following regression models are

specified as follows:4

Performanceit = α0 + α1Divit + α2BHOit + α3MSOit

+α4BSizeit + α5FAgeit + α6Levit

+α7FSizeit + α8NonExecit

+α9Asstplcit + α10Yeardummyit

+α11Industrydummyit + εit (1)

Performanceit = α0 + α1Divpropit + α2BHOit

+α3MSOit + α4BSizeit + α5FAgeit

+α6Levit + α7FSizeit+α8NonExecit

+α9Asstplcit + α10Yeardummyit

+α11Industrydummyit + εit (2) where:

Performance it in equations (1) and (2) stands for ROAit,

ROEit, Tobin’s Qit and CFO/TAit. ROA (Return on assets): Ratio of operating earnings to

total assets. ROE (Return on equity): Ratio of net profit after tax

before abnormals to ordinary shareholders’ equity. CFO/TA (CFO-to-assets): Ratio of cash flow from

operations to total assets. Div (Diversity): Dummy variable equal to 1 in the case

of presence of women on the board, 0 otherwise. Divprop (Diversity proportion): Percentage of women

on the board of directors. BHO (Block-holder ownership): Sum of shares owned

by shareholders who hold more than 5% of a

company’s total shares at the reporting date. MSO (Managerial share-ownership): Percentage of or-

dinary shares owned by directors at the reporting

date. BSize (Board size): Natural logarithm of the number of

directors at the reporting date. FAge (Firm age): Natural logarithm of the firm age. Lev

(Leverage): Ratio of long-term debt to market value

of equity. FSize (Firm size): Natural logarithm of the total assets.

NonExec (Proportion of non-executive directors): Ratio of the total number of non-executive directors to total number of directors.

Asstplc (Assets-in-place): Ratio of the sum of inventory,

property, plant and equipment to total assets.




Descriptive Statistics

Table 2 presents some additional descriptive statistics

indicating the distribution of women on boards in the

sample between 2005 and 2011. The proportion of fe-

male directors to male directors increased from 7.31% in

2005 to 12.26% in 2011 as did the proportion of female

non-executive directors to male non-executive directors,

which increased from 12.42% in 2005 to 18.87% in 2011.

Conversely, the proportion of companies with no female

director on the board decreased from 61.80% to 45.81%

over 2005–2011. Table 2 also shows that the proportion

of female directors on boards increased markedly be-

tween 2010, when ASX amended principles came into

effect, and 2011. Further, following the ASX industry classification,

Table 3 compares the distribution of female directors over

the 10 major industry sectors. Statistics indicate that the

number and percentage of female directors on the boards

of sampled firms varies across the industry sectors over

the sample period. The financials and consumer staples

sectors have the highest number of women directors, with

approximately 11.55% and 10.62% of board members

being female, while women are least likely to be

appointed as directors within the materials (4.05%) and

the energy sectors (4.86%). Table 3 further indicates that

the majority of female directors held non-executive

positions across all industry sectors over 2005–2011. This

ranges from 52.38% of women holding non-executive

positions in the telecommunications services sector to

over 90% of female directors in the financials sector. This

is consistent with the research for the High Pay Centre

and the Guardian, which highlight the lack of women in

executive roles (Neville and Treanor 2012). Table 4 presents a description of the firm

characteristics of the sample. Following prior studies

(e.g., Ahmed et al. 2011), the upper and lower 2% of

three dependent variables including ROA, ROE and

CFO/TA and two independent variables including Asstplc

and Lev, and the upper and lower 10% of Tobin’s Q

(dependent variable) and FSize (independent variable) are

winsorised in order to attenuate the influence of outliers.

The table shows that the mean (median) of ROA, ROE,

Tobin’s Q and CFO/TA are 0.0376 (0.0552), 0.0832

(0.0895), 1.793 (1.404) and 0.0581 (0.0519) respectively.

The standard deviations for these variables do not suggest

significant outliers after winsorisation. The proportion of

female directors ranges from a low of none to a maximum

of 60%, with an average of 6.78%. The percentage of

equity held by block-holders varies between 5% and

100%, with a mean of 39.28%. Managerial share-

ownership ranges from 0% to 99.40%, with an average of

just over 11.25%. The number of directors on boards

ranges from a low of three to a high of 20, with an

average of 7.90. The number of years a company has been

listed on the stock exchange



Pe

rcen

tage

of

co

mp

an

ies w

ith

thre

e o

r m

ore

fe

ma

le d

ire

cto

rs

on

bo

ard

(9)

2.0

1%

2.2

9%

3.3

1%

3.3

8%

1.8

1%

2.4

2%

4.3

4%

Pe

rcen

tage

of

co

mp

an

ies w

ith

two f

em

ale

d

ire

cto

rs

on

bo

ard

(8) 8

.54%

9.6

3%

7.4

6

9.7

7%

11.2

7%

10.3

8%

19.0

6%

Pe

rcen

tage

of

co

mp

an

ies w

ith

on

e f

em

ale

d

ire

cto

r

on

bo

ard

(7) 2

7.6

3%

25.6

8%

27.8

0%

27.4

4%

26.1

8%

29.7

5%

30.7

6%

Pe

rcen

tage

of

co

mp

an

ies w

ith

no

fe

ma

le

dire

cto

r

on

bo

ard

(6) 6

1.8

0%

62.3

8%

61.4

1%

59.3

9%

60.7

2%

57.4

3%

45.8

1%

Pro

po

rtio

n o

f

fem

ale

n

on

-exe

cu

tive

dir

ect

ors

(5) 1

2.4

2%

12.1

8%

12.0

3%

12.4

5%

12.6

0%

13.5

1%

18.8

7%

Pro

po

rtio

n o

f

fem

ale

no

n-

exe

cu

tive

dir

ecto

rs t

o

tota

l n

um

ber

of n

on-

execu

tive d

ire

cto

rs

(4) 8

.79%

8.2

2%

8.1

9%

8.6

1%

8.6

1%

9.4

7%

12

.88

%

sa

mp

le y

ea

rs

Pro

po

rtio

n

of

fem

ale

dire

cto

rs

to t

ota

l n

um

be

r

dire

cto

rs

(3) 5

.88%

5.8

0%

5.9

5%

6.3

3%

6.2

6%

6.9

4%

9.5

1%

in e

ach

of

the

Pro

po

rtio

n

of

fem

ale

to

ma

le

dir

ect

ors

(2) 7

.31%

7.2

0%

7.3

8%

7.8

2%

7.8

5%

9.0

8%

12.2

6%

Bo

ard

div

ers

ity

Num

be

r o

f

com

pa

nie

s

(1) 1

99

218

241

266

275

289

299

Ta

ble

2

Ye

ar

2005

2006

2007

2008

2009

2010

2011

Australian Accounting Review 417


Pro

po

rtio

n o

f fe

ma

le

non-e

xe

cu

tive

dire

cto

rs to

to

tal

nu

mb

er

of

fem

ale

dir

ecto

rs

(3/1

)

88.3

3%

90.6

4%

79.5

6%

75.3

8%

89.5

0%

86.8

4%

78.1

2%

85.5

2%

80%

52.3

8%

Pro

po

rtio

n o

f fe

ma

le

non-e

xe

cu

tive

dire

cto

rs to

to

tal

nu

mb

er

of

no

n-e

xe

cu

tive

dir

ecto

rs

(3/4

)

6.2

2%

17.1

2%

7.9

5%

6.3

3%

13.8

0%

8.8

9%

13.7

3%

14.0

9%

8.1

6%

6.7

0%

Pro

po

rtio

n o

f

fem

ale

dire

cto

rs to

tota

l n

um

be

r o

f d

ire

cto

rs

(1/2

)

4.0

5%

11.5

5%

5.9

1%

4.8

6%

9.3

4%

6%

10.3

2%

10.6

2%

6.6

%

7.5

8%

2005–2011

To

tal n

um

be

r

of n

on

-exe

cu

tive

dir

ect

ors

(4)

17

03

19

80

13

70

774

10

50

371

182

461

392

164

by in

du

str

y s

ec

tor,

To

tal n

um

be

r

of

fem

ale

no

n-

exe

cutiv

e d

ire

cto

rs

(3)

106

339

109

49

145

33

25

65

32

11

Ta

ble

3 N

um

ber

an

d p

erc

en

tag

e o

f fe

ma

le d

ire

cto

rs

To

tal

Tota

l

num

ber

of n

um

ber

of

Industr

y s

ecto

r

fe

ma

le d

irecto

rs d

irecto

rs

(1)

(2

)

Ma

teria

ls

120 2

956

Fin

ancia

ls

374

3

236

Industr

ials

137 2

318

Energ

y 6

5

1335

Consum

er

dis

cre

tio

nary

162 1

734

Health c

are

3

8 633

Info

rma

tio

n t

echnolo

gy 3

2

310

Consum

er

sta

ple

s

76

715

Utilit

ies

40

606

T

ele

com

mu

nic

atio

n s

erv

ices 21

277


ranges from one to 140 years, with an average of over

17 years. The proportion of non-executive directors on

the board ranges from 10% to 90.90%, with an average

of 58.67%. Table 5 presents t-tests of the differences in means for

firms with and without women on their boards. On

examination of the table it is apparent that there are some

significant differences. Firms with female directors are larger

($3 billion in total assets versus $992 million), have larger

boards (an average of nine directors versus almost seven

directors), are more asset-intensive (an average of 0.36

versus 0.33 assets-in-place), and therefore have lower

Tobin’s Q. Firms with female directors have an average age

of over 22 years versus 17 years for firms without women on

their boards. Firms with women on their boards also perform

better, as measured by ROA (0.0666 versus 0.0245) or ROE

(0.136 versus 0.063), and have a higher ratio of CFO/TA

(0.0782 versus 0.0508). Overall, the results indicate that

firms with women on the board are older, larger and perform

better.

Results

Board diversity and firm performance

Table 6 reports the regression estimates on the associ-

ation between board diversity dummy (Div) and firm

performance measures based on the sample consisting of

an unbalanced panel of 1101 firm-year observations for

the period 2005–2011. Consistent with our expectations,

panel OLS regression analyses indicate a positive and

significant association between board diversity and firm

performance. The coefficient (p-value) estimates on Div

for ROA, ROE, TOBINQ and CFO/TA are 0.039 (p <

0.01), 0.0798 (p < 0.01), 0.119 (p < 0.05) and 0.0355 (p <

0.01) respectively. However, endogeneity issues prevent

us from concluding that board diversity causes firms to

perform better. One form of endogeneity problem would

arise if the direction of causality between the dependent

and explanatory variables runs in both directions, the

problem of simultaneity or reverse causality (Brown et al.

2011; Jurkus et al. 2011). The Durbin-Wu Hausman test

is used to investigate the existence of endogeneity. To

perform the test, at the first stage, Div – potential

endogenous variable – is regressed against the control

variables. From this, the residuals are extracted. In the

second stage, the firm performance measures (i.e.,

dependent variables) are regressed against the residuals

obtained in the first stage along with the control

variables. The coefficient (p-value) estimates on Div

residuals for ROA, ROE, TOBINQ and CFO/TA are 0.040 (p < 0.01), 0.08 (p < 0.01), 0.120 (p <

0.05) and 0.036 (p < 0.01) respectively. Significant p-

value estimates on Div residuals across all models

indicate that board diversity variables are endogenous

across all models and therefore OLS regressions are not

418 Australian Accounting Review C 2015 CPA Australia

A. Vafaei, K. Ahmed & P. Mather Board Diversity and Financial Performance

Table 4 Description of firm characteristics

Variable Number of observations Minimum Maximum Mean Median Std. Deviation

ROA 2915 −0.489 0.292 0.0376 0.0552 0.1306

ROE 2884 −0.693 0.7008 0.0832 0.0895 0.226 TOBINQ 2914 0.829 3.865 1.793 1.404 0.984

CFO/TA 2914 −0.317 0.424 0.0581 0.0519 0.125 Divprop 1787 0.000 0.60 0.0678 0.000 0.0928

BHO 2078 0.0500 1.00 0.3928 0.3559 0.223

MSO 2770 0.000 0.9940 0.1125 0.0263 0.174

BSize 1787 3 20 7.90 8 2.751

FAge 3500 1 140 17.024 12 15.871

Lev 2648 0.000 3.909 0.372 0.133 0.719

FSize (Dollar, millions) 2915 72 132 582 8 470 000 000 2 120 000 000 783 000 000 2 740 000 000

ln (FSize) 2915 4.284 9.044 6.679 6.662 1.523

NonExec 1779 0.100 0.9090 0.5867 0.600 0.1413

Asstplc 2364 0.00067 0.877 0.355 0.331 0.271

Table 5 Comparison of the means for firms with and without females on their boards

Variable Average Average t-statistic

(Diversity = 1) (Diversity = 0)

ROA 0.066 0.0245 7.118∗∗∗

ROE 0.136 0.0630 7.241∗∗∗

TOBINQ 1.617 1.952 −7.166∗∗∗

CFO/TA 0.0782 0.0508 4.687∗∗∗

BHO

0.35

0.41 −

5.171∗∗∗

MSO 0.0699 0.155 −10.374∗∗∗

BSize 9.168 6.982 18.017∗∗∗

FAge 22.113 17.064 6.424∗∗∗

Lev 0.388 0.297 2.746∗∗∗

FSize (Dollar, millions) 3 814 685 835 992 037 558 23.967∗∗∗

ln (FSize) 7.575 6.028 24.081∗∗∗

NonExec 0.6317 0.5538 11.919∗∗∗

Asstplc 0.36 0.33 1.781∗ ∗ , ∗∗ and ∗∗∗ are significant at the 0.1, 0.05 and 0.01 levels, respectively.

efficient.5 Gujarati and Porter (2009) recommend that ‘if

the hypothesized equation is determined to actually be

simultaneous, then two-stage least squares (2SLS) or

instrumental variables methods should be used instead of

ordinary least squares’ (Carter et al. 2010: 409). The IV

method requires an instrument that is correlated with

gender diversity but uncorrelated with firm performance

measures, except through the control variables (Adams

and Ferriera 2009; Jurkus et al. 2011). In the corporate

governance context it is usually difficult to find a valid

instrument (Adams and Ferriera 2009; Brown et al. 2011;

Jurkus et al. 2011) because firm characteristics that are

correlated with board diversity are other governance

variables that are already (or should be) included in

performance regressions, such as board size, percentage

of non-executive directors, firm size, etc. (Adams and

Ferriera 2009: 305; Jurkus et al. 2011: 183). Following

Adams and Ferriera (2009: 306) we define our instrument

as ‘the fraction of total board seats in other firms with

female directors’. According to them ‘one reason for

absence of women on boards is their lack of connections’.

As a result, arguably, the more connected

directors are to women, the more female directors should

be observed. Similar to Adams and Ferriera (2009: 306)

we also use variations of this instrument, such as ‘the

fraction of male directors on the board who sit on other

boards on which there are female directors’. Our results are similar using either of the instruments;

however, the correlation of our chosen instrument with

the board diversity variables is higher. Results of the

2SLS tests between board diversity dummy (Div) and

firm performance measures are reported in Table 6. In the

first stage we regress Div against all control variables and

the Instrumental variable (‘Fraction’ from now on). First

stage results indicate a positive and significant

association between Div (dependent variable) and the

Fraction (p < 0.01).6 The Fraction is also uncorrelated

with performance measures except through control

variables included in the regression analysis. This sat-

isfies the second condition for a valid instrument.7 The

results from the 2SLS regression analyses are consistent

with those of OLS regressions; the coefficient on board

diversity is positive and statistically significant across all

models at the 0.01 level. The estimated coefficient


420 A

ustra

lian A

cco

un

ting

Revie

w

C 2

01

5 C

PA

Austra

lia

Table 6 Relation between board diversity (presence of a woman on the board) and firm financial performance

Dep. Var. = ROA Dep. Var. = ROE Dep. Var. = TOBINQ Dep. Var. = CFO/TA

Panel Panel 2SLS Panel Panel 2SLS Panel Panel 2SLS Panel Panel 2SLS

Variable OLS (Second stage) OLS (Second stage) OLS (Second stage) OLS (Second stage)

Intercept 0.0345 0.203 0.136 0.412 3.762 4.494 0.104 0.245

(1.975) (4.20)∗∗∗ (2.634)∗∗∗ (4.52)∗∗∗ (18.118)∗∗∗ (13.58)∗∗∗ (3.329)∗∗∗ (4.75)∗∗∗

Div 0.0396 0.231 0.0798 0.392 0.119 0.950 0.0355 0.195 (4.742)∗∗∗ (6.45)∗∗∗ (5.314)∗∗∗ (5.96)∗∗∗ (1.986)∗∗ (4.02)∗∗∗ (3.911)∗∗∗ (5.10)∗∗∗

BHO 0.0190 0.0620 0.0227 0.0929 −0.309 −0.122 −0.0064 0.0295

(1.0305) (2.70)∗∗∗ (0.684) (2.26)∗∗ (−2.318)∗∗ (−0.83) (−0.3209) (1.20)

MSO 0.0214 0.0359 0.0337 0.0574 0.343 0.406 −0.0217 −0.0095

(0.931) (1.12) (0.815) (1.06) (2.073)∗∗ (2.10)∗∗ (−0.8707) (−0.31)

BSize −0.0451 −0.0844 −0.0501 −0.114 0.0683 −0.102 −0.0079 −0.408

(−3.537)∗∗∗ (−4.31)∗∗∗ (−2.188)∗∗ (−3.46)∗∗∗ (0.744) (−0.88) (−0.5707) (−2.26)∗∗

FAge −0.002 0.00242 −0.0111 −0.00373 −0.0057 0.0138 −0.00438 −0.0006

(−0.408) (0.39) (−1.205) (−0.35) (−0.155) (0.37) (−0.786) (−0.10)

Lev −0.0157 −0.00674 −0.0207 −0.00603 −0.440 −0.401 −0.0229 −0.0153

(−2.394)∗∗ (−0.75) (−1.755)∗ (−0.46) (−9.291)∗∗∗ (−7.31)∗∗∗ (−3.201)∗∗∗ (−1.64)

FSize 0.0179 −0.0019 0.0188 −0.0137 −0.253 −0.339 0.00324 −0.134

(5.347)∗∗∗ (−0.37) (3.117)∗∗∗ (−1.40) (−10.460)∗∗∗ (−9.90)∗∗∗ (0.889) (−2.38)∗∗

NonExec −0.0449 −0.171 −0.0930 −0.3005 −0.103 −0.653 −0.0623 −0.168

(−1.592) (−3.86)∗∗∗ (−1.835)∗ (−3.88)∗∗∗ (−0.508) (−2.32)∗∗ (−2.033)∗∗ (−3.54)∗∗∗

Asstplc 0.0185 0.0482 0.0106 0.0591 −0.296 −0.168 0.0518 0.0766

(1.369) (2.91)∗∗∗ (0.437) (2.04)∗∗ (−3.039)∗∗∗ (−1.59) (3.519)∗∗∗ (4.84)∗∗∗

Number of 224 224 224 224 224 224 224 224

cross-sections

Number of 1101 1101 1101 1101 1101 1101 1101 1101

observations

(unbalanced

panel)

R2

0.06435 0.0793 0.04511 0.0554 0.2626 0.1340 0.0394 0.0542 Adjusted R

2 0.0566 0.0658 0.0372 0.0483 0.2565 0.1268 0.0315 0.0513

F-statistic 8.33∗∗∗ 12.913∗∗∗ 5.727∗∗∗ 8.723∗∗∗ 43.181∗∗∗ 45.550∗∗∗ 4.976∗∗∗ 7.455∗∗∗

∗ , ∗∗ and ∗∗∗ are significant at the 0.1, 0.05 and 0.01 levels, respectively with one-tailed significance applied where the directionality of the sign of the coefficient was predicted and two-

tailed where directionality was not specified. Coefficient values are shown outside the parentheses, with t-stat values shown inside the parentheses. The different variables used in the table are defined as follows: Div = Dummy variable equal to 1 in case of the presence of a woman on the board, 0 otherwise. BHO = Sum of shares owned by

shareholders who hold more than 5% of a company’s total shares at the reporting date. MSO = Percentage of ordinary shares owned by directors at the reporting date. BSize = Natural logarithm of the

number of directors at the reporting date. FAge = Natural logarithm of firm age. Lev = Ratio of book value of long-term debt by market value of equity. FSize = Natural logarithm of total assets. NonExec

= proportion of non-executive directors on board compared to total number of directors. Asstplc = Ratio of the sum of inventory, property, plant and equipment to total assets.

Board

Div

ers

ity a

nd F

inancia

l Pe

rform

ance

A. V

afa

ei, K

. Ahm

ed &

P. M

ath

er


of board diversity suggests that a 1% increase in

female directors is associated with a 0.231% increase

in ROA, 0.392% increase in ROE, 0.950% increase in

TOBINQ and 0.195% increase in CFO/TA. With respect to control variables, BSize indicates a sig-

nificant and negative association with all performance

measures – except for TOBINQ. This is consistent with

the argument that larger boards are less likely to function

effectively due to coordination and communication

problems, which would negatively affect board effective-

ness and also firm performance (Kameda et al. 1992;

Lipton and Lorsch 1992; Jensen 1993). The direction of

association between leverage and all firm performance

measures is negative but leverage has only a significant

and negative association with TOBINQ. This is consis-

tent with the results of Opler and Titman (1994) and

Gonzalez´ (2013) and indicates that market performance

of firms with higher leverage is significantly reduced.

Asset intensity has a positive and significant association

with accounting and economic-based measures of firm

performance. This indicates that higher asset intensity is

associated with improved accounting and economic per-

formance. FSize has a negative and significant association

with TOBINQ and CFO/TA. The direction of association

between FSize and other firm performance measures is

negative but insignificant. This suggests that not nec-

essarily larger firms perform better (Hart and Oulton

1996), and in fact smaller firms may have an incentive to

grow with the prospect of exploiting various bene-fits

attributed to larger firms (Symeou 2010: 3). Other results

reported in Table 6 indicate that MSO is only positively

and significantly associated with TOBINQ and it does not

have a significant association with other firm performance

measures. This is consistent with the findings of Craswell

et al. (1997) who indicate that MSO is not an important

determinant of Australian corporate performance.8 BHO

indicates a positive association with ROA and ROE and a

negative and significant association with TOBINQ. Such

inconsistent results could be due to nonlinear association

of BHO with firm performance (Thomsen et al. 2006).

Finally NonExec indicates a negative and significant

association with all performance measures.9 All models

are highly significant, with F-statistic ranging between

7.45 and 45.55, although adjusted R2 ranges between

4.83% and 12.68%. Table 7 reports the regression estimates on the asso-

ciation between proportion of female board members

(Divprop) and firm performance measures. OLS regres-

sion results indicate a positive and significant association

between Divprop and all firm performance measures.

However, due to the existence of endogeneity10

2SLS re-

gressions are conducted. The first stage of the regression

analysis indicates that the Fraction is significantly and

positively associated with Divprop (p < 0.01).11

The sec-

ond stage regression coefficient estimates on Divprop are

positive and significant across all four models (p < 0.01).


These are consistent with the results reported in Table 6

and once again support our hypothesis. The control vari-

ables in Table 7 indicate a relatively similar pattern with

Table 6 variables. The models are significant with com-

parable adjusted R2 as reported in Table 6.

Overall, the positive association between firm perfor-

mance and board diversity is consistent with theoretical

predictions underpinning resource dependence and

agency perspectives. The results suggest that bringing to-

gether a diverse range of skills and experience on boards

improves business performance and could improve deci-

sion effectiveness (Australian Institute of Company Di-

rectors 2013). In fact, as stated by John Colvin, Chief

Executive Officer and Managing Director of the Aus-

tralian Institute of Company Directors, ‘It is not only

good social policy but it makes good business sense that

our boards and companies benefit from all the talents that

is available to them’ (Australian Institute of Company

Directors 2013).

Further analysis

Owing to the negative association between the percent-

age of non-executive directors and firm performance

measures, we investigate the association between the

percentage of female non-executive directors (FEM-

NONEXEC) and firm performance. Table 8 reports the

regression estimates on the association between

percentage of female non-executive directors and firm

performance measures. The correlation coefficient

between FEMNONEXEC and Div (Divprop) is 0.811

(0.917) at p < 0.01. As a result Div (Divprop) is excluded

from the regression analysis. Due to the existence of

endogeneity 2SLS tests are conducted. The first-stage

results indicate a positive and significant association

between FEMNONEXEC and the Fraction (p < 0.01).12

The second-stage results indicate a significant and

positive association between FEMNONEXEC and all firm

performance measures. The association between NonExec

and all firm performance measures remains negative and

significant. Results remain the same by using variations

of the instrument. We also examine the effect of year and industry

dummies and the interaction of industry dummy and

board diversity on firm performance measures. Untab-

ulated results show that the coefficient on Div (Divprop)

remains significant across all models; however, neither

the year dummy or industry sector dummy variables, nor

the interaction of industry sector and board diversity

indicators (including the dummy variable and proportion

of female directors) are found to have a significant

relationship with firm performance. We also investigate

the association between squared Divprop and firm

performance measures to test the proposition that a

critical mass of women is required for board


422 A

ustra

lian A

cco

un

ting

Revie

w

C 2

01

5 C

PA

Austra

lia

Table 7 Relation between board diversity (proportion of women on the board) and firm financial performance


Panel Panel 2SLS Panel Panel 2SLS Panel Panel 2SLS Panel Panel 2SLS

Variable OLS (Second stage) OLS (Second stage) OLS (Second stage) OLS (Second stage)

Intercept 0.0149 0.0828 0.0981 0.207 3.687 3.999 0.0848 0.143

(0.531) (2.18)∗∗ (1.945)∗ (2.95)∗∗∗ (18.215)∗∗∗ (15.28)∗∗∗ (2.774)∗∗∗ (3.52)∗∗∗

Divprop 0.195 1.0609 0.407 1.802 0.386 4.358 0.1502 0.898

(4.575)∗∗∗ (6.30)∗∗∗ (5.317)∗∗∗ (5.89)∗∗∗ (1.924)∗ (3.98)∗∗∗ (3.237)∗∗∗ (4.92)∗∗∗

BHO 0.0195 0.0609 0.0243 0.0911 −0.317 −0.127 −0.00723 0.0285

(1.052) (2.86)∗∗∗ (0.731) (2.39)∗∗ (−2.376)∗∗ (−0.88) (−0.358) (1.20)

MSO 0.0205 0.0303 0.0322 0.0479 0.338 0.383 −0.0227 −0.0143

(0.895) (0.99) (0.7803) (0.94) (2.042)∗∗ (2.06)∗∗ (−0.909) (−0.48)

BSize −0.0392 −0.0494 −0.0385 −0.0549 0.0883 0.0417 −0.00238 −0.0111

(−3.103)∗∗∗ (−2.90)∗∗∗ (−1.696)∗ (−1.90)∗∗ (0.969) (0.40) (−0.173) (−0.70)

FAge −0.0018 0.00365 −0.0104 −0.00163 −0.0061 0.0189 −0.00427 0.000445

(−0.3507) (0.60) (−1.130) (−0.16) (−0.165) (0.51) (−0.764) (0.08)

Lev −0.0164 −0.0108 −0.0219 −0.01309 −0.444 −0.418 −0.0236 −0.0188

(−2.489)∗∗ (−1.56) (−1.856)∗ (−1.23) (−9.357)∗∗∗ (−7.90)∗∗∗ (−3.298)∗∗∗ (−2.32)∗∗

FSize 0.0191 0.00607 0.0209 −0.00007 −0.246 −0.306 0.00486 −0.00662

(5.784)∗∗∗ (1.25) (3.530)∗∗∗ (−0.01) (−10.334)∗∗∗ (−10.65)∗∗∗ (1.298) (−1.33)

NonExec −0.0394 −0.131 −0.0834 −0.231 −0.0652 −0.487 −0.0547 −0.134

(−1.405) (−3.19)∗∗∗ (−1.657)∗ (−3.27)∗∗∗ (−0.322) (−1.84)∗ (−1.794)∗∗ (−3.00)∗∗∗

Asstplc 0.01709 0.0378 0.0080 0.0414 −0.305 −0.2109 0.0499 0.0678

(1.262) (2.44)∗∗ (0.330) (1.52) (−3.136)∗∗∗ (−2.06)∗∗ (3.388)∗∗∗ (4.56)∗∗∗

Number of 224 224 224 224 224 224 224 224

cross-sections

Number of 1101 1101 1101 1101 1101 1101 1101 1101

observations

(unbalanced panel)

R2

0.06305 0.0781 0.0451 0.0545 0.261 0.1477 0.0352 0.0513 Adjusted R

2 0.0553 0.0674 0.0372 0.0474 0.254 0.1407 0.0272 0.0504

F-statistic 8.158∗∗∗ 12.918∗∗∗ 5.731∗∗∗ 8.723∗∗∗ 42.826∗∗∗ 45.550∗∗∗ 4.426∗∗∗ 7.455∗∗∗

∗ ,∗∗ and ∗∗∗ represent significance at the 0.1, 0.05 and 0.01 levels, respectively with one-tailed significance applied where the directionality of the sign of the coefficient was predicted, and

two-tailed where directionality was not specified. Coefficient values are shown outside the parentheses, with t-stat values shown inside the parentheses. The different variables used in the table are defined as follows: Divprop = Proportion of female directors on the board compared to the total number of directors. BHO = Sum of shares owned by

shareholders who hold more than 5% of a company’s total shares at the reporting date. MSO = Percentage of ordinary shares owned by directors at the reporting date. BSize = Natural logarithm of the

number of directors at the reporting date. FAge = Natural logarithm of firm age. Lev = Ratio of book value of long-term debt by market value of equity. FSize = Natural logarithm of total assets. NonExec

= proportion of non-executive directors on board compared to total number of directors. Asstplc = Ratio of the sum of inventory, and property, plant and equipment to total assets.

Board

Div

ers

ity a

nd F

inancia

l Pe

rform

ance

A. V

afa

ei, K

. Ahm

ed &

P. M

ath

er

A. Vafaei, K. Ahmed & P. Mather Board Diversity and Financial Performance

Table 8 Relation between female non-executive directors and firm financial performance


Variable Panel 2SLS Panel 2SLS Panel 2SLS Panel 2SLS

(Second stage) (Second stage) (Second stage) (Second stage)

Intercept 0.0964 0.230 4.055 0.155

BHO

(2.7.02)∗∗∗ (3.682)∗∗∗ (17.172)∗∗∗ (4.211)∗∗∗

0.0509 0.0740 −0.168 0.0200

(2.275)∗∗

(1.886)∗ ( −

(0.869)

MSO

1.139)

0.0393 0.0632 0.420 −0.006

(1.455)

(1.332)

(2.350)∗∗ ( −

BSize

0.240)

−0.0517 −0.0588 0.0322 −0.0131

(

−

3.466)∗∗∗ ( −

(0.326)

( −

2.246)∗∗ 0.852)

FAge 0.0036 −0.00163 0.0189 0.0004

Lev

(0.602) (−0.152) (0.471) (0.0711)

−0.0046 −0.00247 −0.393 −0.0135

(

−

0.587) ( −

( −

( −

0.177) 7.495)∗∗∗ 1.661)∗

FSize 0.00803 0.00324 −0.298 −0.0049

(1.848)∗

(0.425) −

( −

NonExec

( 10.383)∗∗∗ 1.108)

−0.160 ( −0.281 ( −0.607 ( −0.159 (

− 4.142 )

∗∗∗ 4.133 )

∗∗∗ 2.368 )

∗∗ 3.975 )

∗∗∗

FEMNONEXEC − − −

1.179 2.003 4.845 0.999

Asstplc

(6.617)∗∗∗ (6.405)∗∗∗ (4.110)∗∗∗ (5.434)∗∗∗

0.0369 0.0400 −0.214 0.0671

(2.283)∗∗

(1.407) (

−

(4.019)∗∗∗

Number of cross-sections 2.001)∗∗

224 224 224 224

Number of observations (unbalanced panel) 1101 1101 1101 1101

R2 0.0487 0.0452 0.1514 0.0291

Adjusted R2

0.0476 0.0424 0.1444 0.0283

F-statistic 12.918∗∗∗ 8.723∗∗∗ 45.550∗∗∗ 7.455∗∗∗ ∗ , ∗∗ and ∗∗∗ represent significance at the 0.1, 0.05 and 0.01 levels, respectively with one-tailed significance applied where the

directionality of the sign of the coefficient was predicted and two-tailed where directionality was not specified. Coefficient values are

shown outside the parentheses, with t-stat values shown inside the parentheses. The different variables used in the table are defined as follows: Div = Dummy variable equal to 1 in case of the presence of women

on the board, 0 otherwise. BHO = Sum of shares owned by shareholders who hold more than 5% of a company’s total shares at

the reporting date. MSO = Percentage of ordinary shares owned by directors at the reporting date. BSize = Natural logarithm of the

number of directors at the reporting date. FAge = Natural logarithm of firm age. Lev = Ratio of book value of long-term debt by

market value of equity. FSize = Natural logarithm of total assets. NonExec = proportion of non-executive directors on board

compared to total number of directors. FEMNONEXEC = proportion of female non-executives on board compared to total number

of directors. Asstplc = Ratio of the sum of inventory, property, plant and equipment to total assets.

diversity to have a significant influence on firm perfor-

mance (Joecks et al. 2013). The coefficient estimates on

squared Divprop are not significant for either of the firm

performance measures.13

This indicates that the rela-

tionship between Div (Divprop) and firm performance is a

linear one. This is in contrast to the finding of Joecks et

al. (2013) who find a non-linear association between

board diversity and firm performance measures. We also

control for the effect of other corporate governance

variables such as CEO duality. However, we did not

include this variable in the main analysis because the

number of firms in our sample with CEO duality is quite

low, which would have generated a less reliable

coefficient. The further analysis that controls for this

variable gives results that are qualitatively similar.

Conclusion

This paper examines the association between board diversity and firm performance. Gender diversity (as C 2015 CPA Australia

part of board diversity) is a topical research issue to

assess whether gender balance adds value to corporate

boards. It is argued that this added value is in

maximising shareholders’ wealth and bringing more

‘equity and justice’ into boards and corporations. Two

theoretical perspectives, agency and resource

dependence theory, that underpin this research suggest

that there should be a positive association between

board diversity and firm performance.

The policy motivation for this study arises from the

amendments made to the ASX Corporate Governance

Council principles and recommendations implicitly re-

quiring companies to establish policies to increase board

diversity. The proportion of women who reach senior

positions in the business sector remains very low in

Australia. If it could be demonstrated that on aver-age,

women on boards of directors enhance firm per-

formance, then there is a business case to support the

regulatory pressure to appoint more women on boards.

Using a sample of the top 500 ASX listed firms over

the period between 2005 and 2011, this study empirically



analysed whether the proportion of women on boards of

directors affects firm performance. The results indicate

that there is a trend toward the appointment of women on

corporate boards in Australia since 2010, and that it is

mainly the older, larger and better performing firms that

have female representatives on their boards. Also, despite

the typically low percentage of female directors, this

study finds a positive association between board diversity

and firm financial performance after control-ling for

several firm-specific and governance variables. Further

tests for the causality between the proportion of women

on boards of directors and financial performance suggest

that it is board diversity that affects performance, not the

opposite. A potential limitation is that a sample of the top

500 ASX listed firms means that the findings of this study

may not be generalisable to smaller ASX listed firms.

This study has implications for practice and regu-

lation. Showing that women on boards is associated with

enhanced firm performance supports the business case for

the recent Australian regulatory change, informs board

nomination committees and complements social justice

arguments for increasing the number of women on

boards. This study also highlights possibilities for future

research. Future work could investigate the effect of

board diversity and its interaction with various factors

such as tenure and expertise on firm performance. There

is also scope for qualitative research, such as interviews

with board members and corporate directors, on per-

ceptions of female directors’ impact on monitoring and

strategic decisions.

Acknowledgements

Earlier versions of this paper were presented at the 2013

Accounting and Finance Association of Australia and

New Zealand (AFAANZ) Conference, the 3rd

Financial

Markets and Corporate Governance Conference held in

Melbourne in April 2012, as well as La Trobe University,

Department of Accounting Seminar Series (semester 1,

2012). The authors would like to gratefully acknowledge

helpful comments and suggestions from Darren Henry,

John Hillier, Donald Stokes, Robert Faff, Zahirul Hoque,

Xu Dong Ji, Jahangir Ali, Judy Louie and Biserka Siladi.

In addition, financial assistance received via a research

grant from AFAANZ, as well as an internal research grant

received from the Faculty of Business, Economics and

Law, La Trobe University, is gratefully acknowledged.

Notes

1 Board diversity relates to the degree of similarity and difference

between the directors on a board. There are a number of ways to

describe differences between individuals, which may include


‘measurable factors such as age, gender, and socio-economic or

cultural background’ (Corporations and Markets Advisory

Committee (CAMAC) 2009: 23). 2 As indicated by Givoly and Hayn (2000: 300), ‘a measure of

individual company performance which is unaffected by

accrual accounting is the cash flow from operations’.

Following them, this study uses the ratio of CFO-to-assets to

assess firms’ economic performance. 3 Estimates of the replacement costs of assets are not reported in

Australia. 4 In the models, we also control for the effect of year and

industry dummies and the interaction of industry dummies and

board diversity. 5 The full results of these tests include the regression results for

two regression equations for each model and a total of eight

regression equations across all models. Therefore, we do not

report all the results to save space. Full results are available

upon request from the authors. 6 The first-stage model between Div and the Fraction is highly

significant with F-statistic at 81.12 and adjusted R2 at 29.62%.

7 Full results of the first-stage regression analyses are available upon request.

8 According to Craswell et al. (1997) and Khan et al. (2014) the

association between MSO and firm performance could be a

non-linear one. 9 Further tests are conducted to investigate the direction of

association between proportion of female non-executive

directors and firm performance measures. 10 The coefficient (p-value) estimates on Divprop residuals for

ROA, ROE, TOBINQ and CFO/TA are 0.195 (p < 0.01), 0.407

(p < 0.01), 0.386 (p < 0.1) and 0.150 (p < 0.01) respectively.

The full set of results is available upon request. 11 The first-stage model between Divprop and the Fraction is highly

significant with F-statistic at 39.17 and adjusted R2 at 21.08%.

12 The first-stage model between FEMNONEXEC and the Fraction is

highly significant with F-statistic at 36.762 and adjusted R2 at

22.63%. First-stage results are available upon request. 13 We test the association between squared Divprop and firm per-

formance measures by using 2SLS tests. To do this, in the

second stage, squared Divprop along with Divprop and other

control variables are regressed against all firm performance

measures. To run this set of 2SLS tests, we use two different

instrumental variables including ‘the fraction of total board

seats in other firms with female directors’ and ‘the fraction of

male directors on the board who sit on other boards on which

there are female directors’.

References

Adams, R.B. and Ferreira, D. 2009, ‘Women in the Boardroom

and Their Impact on Governance and Performance’, Journal of

Financial Economics, 94 (2): 291–309. Agrawal, A. and Knoeber, C.R. 1996, ‘Firm Performance

and Mechanisms to Control Agency Problems between

Managers and Shareholders’, The Journal of Financial and

Quantitative Analysis, 31 (3): 377–97. Ahmed, K., Hillier, J. and Tanusasmita, E. 2011, ‘R&D

Profitability, Intensity and Market-to-book: Evidence from

Australia’, Accounting Research Journal, 24 (2): 150–77. Ahren, K.R. and Dittmar, A. 2012, ‘The Changing of the Boards:

The Impact on Firm Valuation of Mandated Female Board




Representation’, The Quarterly Journal of Economics, 127: 137– 97. Anderson, R.C. and Reeb, D.M. 2003, ‘Founding-family

Ownership and Firm Performance: Evidence from the S&P

500’, The Journal of Finance, 58 (3): 1301–27. Australian Institute of Company Directors 2009, ‘Directors

welcome new ASX guidelines on diversity’, Media Release,

7 December. Australian Institute of Company Directors 2013, ‘Board diver-

sity – Think outside the square’, Media Release, 30 October. ASX Corporate Governance Council 2010, ‘Proposed Amend-

ments to the ASX Corporate Governance Council Corporate

Governance Principles and Recommendations’. Available at:

http://www.asx.com.au/about/pdf/20100421_proposed_

changes_to_corporate_governance_principles.pdf, accessed 11

March 2011. Baker, G.P. and Kennedy, R.E. 2002, ‘Survivorship and the

Economic Grim Reaper’, Journal of Law, Economics, and

Organization, 18 (2): 324–61. Bantel, K.A. and Jackson S.E. 1989, ‘Top Management and

Innovations in Banking: Does the Composition of the Top

Team Make a Difference?’, Strategic Management Journal,

10: 107–24. Bonn, I. 2004, ‘Board Structure and Firm Performance: Evi-

dence from Australia’, Journal of Management &

Organization, 10 (1): 14–24. Brooke, J.K. and Tyler, T.R. 2011, ‘Board Diversity and

Corporate Performance: Filling in the Gaps: Diversity and

Corporate Performance: A Review of the Psychological

Literature’, North Carolina Law Review, 89: 715–1055. Brown, P., Beekes, W. and Verhoeven, P. 2011, ‘Corporate

Governance, Accounting and Finance: A Review’,

Accounting and Finance, 51: 96–172. Campbell, K. and Minguez-Vera, A. 2008, ‘Gender Diversity

in the Boardroom and Firm Financial Performance’, Journal

of Business Ethics, 83 (3): 435–51. Carter, D.A., Simkins, B.J. and Simpson, W.G. 2003,

‘Corporate Governance, Board Diversity, and Firm Value’,

Financial Review, 38 (1): 33–53. Carter, D.A., D’Souza, F., Simkins, B.J. and Simpson, W.G.

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. Corporations and Market Advisory Committee (CAMAC)

2009, ‘Diversity on Boards of Directors’. Available at:

www.camac.gov.au, accessed 22 August 2010. Cox, T.H. and Blake, S. 1991, ‘Managing Cultural Diversity:

Implications for Organizational Competitiveness’, The

Executive, 5 (3): 45–56. Craswell, A.T., Taylor, S.L. and Saywell, R.A. 1997,

‘Ownership Structure and Corporate Performance’, Pacific-

Basin Finance Journal, 5: 301–23.


Dale-Olsen, H., Schøne, P. and Verner, M. 2013, ‘Diversity

among Norwegian Boards of Directors: Does a Quota for

Women Improve Firm Performance?’, Feminist Economics,

19 (4): 110–35. Dalton, D.R., Daily, C.M., Johnson, J.L. and Ellstrand, A.E.

1999, ‘Number of Directors and Financial Performance: A

Meta-analysis’, Academy of Management Journal, 42 (6):

674– 86. Dunstan, K., Keeper, T., Truong, T.P. and van Zijl, T. 2011,

‘The Influence of Board Gender Diversity on the Value of

NZX Listed Firms and its Association with Growth

Options’, Working Paper, Bond University. Eklund, J., Palmgren, J. and Wiberg, D. 2009, ‘Ownership

Structure, Board Composition, and Investment Performance’,

Working Paper, The Ratio Institute, Stockholm. Erhardt, N.L., Werbel, J.D. and Shrader, C.B. 2003, ‘Board of Director Diversity and Firm Financial Performance’, Corporate Governance: An International Review, 11 (2): 102–11. Farrell, K.A. and Hersch, P.L. 2005, ‘Additions to Corporate

Boards: The Effect of Gender’, Journal of Corporate

Finance, 11 (1): 85–106. Ferreira, D. 2010, ‘Board Diversity’, in Anderson R. and

Baker, H.K. (Eds), Corporate Governance: A Synthesis of

Theory, Research, And Practice, John Wiley and Sons, New

York: 225–42. Frye, M. 2004, ‘Equity-based Compensation for Employees:

Firm Performance and Determinants’, Journal of Financial

Re-search, 27 (1): 31–54. Givoly, D. and Hayn, C. 2000, ‘The Changing Time-series

Properties of Earnings, Cash Flows and Accruals: Has

Financial Re-porting become More Conservative?’, Journal

of Accounting and Economics, 29: 287–320. Gonzalez,´ V.M. 2013, ‘Leverage and Corporate

Performance: International Evidence’, The International

Review of Economics and Finance, 25: 169–84. Gooding, R.Z. and Wagner III, J.A. 1985, ‘A Meta-analytic

Review of the Relationship between Size and Performance:

The Productivity and Efficiency of Organizations and their

Subunits’, Administrative Science Quarterly, 30 (4): 462–81.

Goodstein, J., Gautam, K. and Boeker, W. 1994, ‘The

Effects of Board Size and Diversity on Strategic Change’,

Strategic Management Journal, 15 (3): 241–50. Gujarati, D.N. and Porter, D.C. 2009, Basic Econometrics,

5th

edn, McGraw-Hill, New York. Gul, F.A., Hutchinson, M. and Lai, K.M.Y. 2013, ‘Gender-

diverse Boards and Properties of Analyst Earnings Forecasts’,

Accounting Horizons, 27 (3): 511–38. Gul, F.A., Srinidhi, B. and Ng, A.C. 2011, ‘Does Board

Gender Diversity Improve the Informativeness of Stock

Prices?’, Journal of Accounting and Economics, 51: 314–38.



Hart, O. 1995, ‘Corporate Governance: Some Theory and Implications’, Economic Journal, 105: 678–89. Hart, P.E. and Oulton, N. 1996, ‘Growth and Size of Firms’, The Economic Journal, 106 (438): 1242–52. Haslam, S.A., Ryan, M.K., Kulich, C., Trojanowski, G. and

Atkins, C. 2010, ‘Investing with Prejudice: The Relationship

between Women’s Presence on Company Boards and Objective

and Subjective Measures of Company Performance’, British

Journal of Management, 21 (2): 484–97. Healy, P.M., Palepu, K.G. and Ruback, R.S. 1992, ‘Does

Corporate Performance Improve after Mergers?’, Journal of

Financial Economics, 31 (2): 135–75. Henry, D. 2008, ‘Corporate Governance Structure and the Valuation

of Australian Firms: Is There Value in Ticking the Boxes’,

Journal of Business Finance and Accounting, 35 (7–8): 912–42. Hillman, A.J., Cannella, J.A.A. and Paetzold, R.L. 2000, ‘The

Resource Dependence Role of Corporate Directors: Strategic

Adaption of Board Composition in Response to Environmental

Change’, Journal of Management Studies, 37: 235–55. Jensen, M.C. 1993, ‘The Modern Industrial Revolution, Exit,

and the Failure of Internal Control Systems’, The Journal of

Finance, 48 (3): 831–80. Joecks, J., Pull, K. and Vetter, K. 2013, ‘Gender Diversity in the

Boardroom and Firm Performance: What Exactly Constitutes a

“Critical Mass?”’, Journal of Business Ethics, 118 (1): 61–72. Jurkus, A.F., Park, J.C. and Woodard, L.S. 2011, ‘Women in

Top Management and Agency Costs’, Journal of Business

Research, 64 (2): 180–86. Kameda, T., Stasson, M.F., Davis, J.H., Parks, C.D. and

Zimmer-man, S.K. 1992, ‘Social Dilemmas, Subgroups, and

Motivation Loss in Task-oriented Groups: In Search of an

“Optimal” Team Size in Division of Work’, Social

Psychology Quarterly, 55 (1): 47–56. Khan, A., Mather, P. and Balachandran, B. 2014, ‘Managerial

Share Ownership and Operating Performance: Do Independent

and Executive Directors Have Different Incentives?’, Aus-

tralian Journal of Management, 39 (1): 47–71. Korporaal, G. 2010, ‘Women Make Mark on Board’, The Australian, Business section, August: 37. Krishnan, H. and Park, D. 2005, ‘A Few Good Women – On

Top Management Teams’, Journal of Business Research, 58

(12): 1712–20. Lee, P.M. and James, E.H. 2007, ‘She’-e-os: Gender Effects

and Investor Reactions to the Announcements of Top

Executive Appointments’, Strategic Management Journal,

28 (3): 227–41. Lipton, M. and Lorsch, J. 1992, ‘A Modest Proposal for Im-

proved Corporate Governance’, The Business Lawyer, 48 (1):

59–77. Luckerath-Rovers, M. 2013, ‘Women on Boards and Firm

Performance’, Journal of Management and Governance, 17

(2): 491– 509.


Majumder, S.K. 1997, ‘The Impact of Size and Age on Firm-

level Performance: Some Evidence from India’, Review of

Industrial Organization, 12 (2): 231–41. Mahadeo, J.D., Soobaroyen, T. and Hanuman, V.O. 2012,

‘Board Composition and Financial Performance: Uncovering

the Effects of Diversity in an Emerging Economy’, Journal

of Business Ethics, 105: 375–88. McCann, M. and Wheeler, S. 2011, ‘Gender Diversity in the

FTSE 100: The Business Case Claim Explored’, Journal of

Law and Society, 38 (4): 542–74. Miller, T. and Triana, M.D.C. 2009, ‘Demographic Diversity

in the Boardroom: Mediators of the Board Diversity – Firm

Performance Relationship’, Journal of Management Studies,

46 (5): 755–86. Mohan, N.J. and Chen, C.R. 2004, ‘Are IPOs Priced Differ-

ently Based on Gender?’, Journal of Behavioral Finance, 5

(1): 57–65. Moncrief, W.C., Babakus, E., Cravens, D.W. and Johnston,

M.W. 2000, ‘Examining Gender Differences in Field Sales

Organizations’, Journal of Business Research, 49 (3): 245–

57. Morck, R., Shleifer, A. and Vishny, R.W. 1988, ‘Management

Ownership and Market Valuation: An Empirical Analysis’,

Journal of Financial Economics, 20: 293–315. Murray, A.I. 1989, ‘Top Management Group Performance and

Firm Performance’, Strategic Management Journal, 20: 125–41. Neville, S. and Treanor, J. 2012, ‘Research Highlights Lack

of Women in Executive Roles’, The Guardian, 12

December. Available at: http://www.theguardian.com/

business/2012/dec/12/research-lack-women-executive-roles,

accessed 10 December 2013. Nguyen, H. and Faff, R. 2007, ‘Impact of Board Size and

Board Diversity on Firm Value: Australian Evidence’,

Corporate Ownership & Control, 4 (2): 24–32. Opler, T.C. and Titman, S. 1994, ‘Financial Distress and

Corporate Performance’, The Journal of Finance, 94 (3):

1015–40. Pearce II, J.A. and Zahra, S.A. 1992, ‘Board Composition

from a Strategic Contingency Perspective’, Journal of

Management Studies, 29 (4): 411–38. Pfeffer, J. 1973, ‘Size, Composition, and Function of

Hospital Boards of Directors: A Study of Organization–

Environment Linkage’, Administrative Science Quarterly, 18

(3): 349–64. Robinson, G. and Dechant, K. 1997, ‘Building a Business

Case for Diversity’, The Academy of Management Executive

(1993– 2005), 11 (3): 21–31. Rose, C. 2007, ‘Does Female Board Representation Influence

Firm Performance? The Danish Evidence’, Corporate Gover-

nance: An International Review, 15 (2): 404–13. Rubin, M. 2000, ‘Women in the American Society for Public Administration: Another Decade of Progress but

426 Australian Accounting Review C 2015 CPA Australia


Still a Way to Go’, Public Administration Review, 60 (1): 61–71. Short, H. and Keasey, K. 1999, ‘Managerial Ownership and

the Performance of Firms: Evidence from the UK’, Journal

of Corporate Finance, 5 (1): 79–101. Shrader, C.B., Blackburn, V.B. and Iles, P. 1997, ‘Women in

Management and Firm Financial Performance: An Ex-

ploratory Study’, Journal of Managerial Issues, 9 (3): 355–

72. Smith, N., Smith, V. and Verner, M. 2006, ‘Do Women in

Top Management Affect Firm Performance? A Panel Study

of 2,500 Danish Firms’, International Journal of

Productivity and Performance Management, 55 (7): 569–93. Standard and Poor’s 2004, ‘Understanding Indices’.

Available at: http://www.standardandpoors.com/home/en/us,

accessed 11 March 2012.


Symeou, P.C. 2010, ‘The Firm Size–Performance Rela-

tionship: An Empirical Examination of the Role of the

Firm’s Growth Potential’, Working Paper, University of

Munich. Thomsen, S., Pedersen, T. and Kvist, H.K. 2006,

‘Blockholder Ownership: Effects on Firm Value in Market

and Control Based Governance Systems’, Journal of

Corporate Finance, 12 (2): 246– 69. Wang, Y. and Clift, B. 2009, ‘Is There a “Business Case” for

Board Diversity?’, Pacific Accounting Review, 21 (2): 88–

103. Wruck, K.H. 1990, ‘Financial Distress, Reorganization, and

Organizational Efficiency’, Journal of Financial Economics,

27 (2): 419–44. Zingales, L. 2000, ‘In Search of New Foundations’, The Journal of Finance, 55 (4): 1623–53.


board diversity and financial performance in the top 500 ... · board diversity and financial...

Documents