Description
As a result of the Australian Government Productivity Commission’s recommendation to
mandate remuneration committee independence for ASX300 companies, this study aims to investigate
whether voluntary remuneration committee independence aligns chief executive officer (CEO) total
pay and bonuses with firm financial performance.
Accounting Research Journal
Remuneration committee independence and CEO remuneration for firm financial
performance
Patti Cybinski Carolyn Windsor
Article information:
To cite this document:
Patti Cybinski Carolyn Windsor , (2013),"Remuneration committee independence and CEO remuneration
for firm financial performance", Accounting Research J ournal, Vol. 26 Iss 3 pp. 197 - 221
Permanent link to this document:
http://dx.doi.org/10.1108/ARJ -08-2012-0068
Downloaded on: 24 January 2016, At: 21:18 (PT)
References: this document contains references to 63 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1807 times since 2013*
Users who downloaded this article also downloaded:
Victoria J . Clout, Larelle Chapple, Nilan Gandhi, (2013),"The impact of auditor independence regulations
on established and emerging firms", Accounting Research J ournal, Vol. 26 Iss 2 pp. 88-108 http://
dx.doi.org/10.1108/ARJ -Dec-2011-0045
Karol Marek Klimczak, Grzegorz Szafranski, (2013),"Coincident and forecast relevance of accounting
numbers", Accounting Research J ournal, Vol. 26 Iss 3 pp. 239-255 http://dx.doi.org/10.1108/
ARJ -09-2012-0076
Prodosh Simlai, (2014),"Firm characteristics, distress risk and average stock returns", Accounting Research
J ournal, Vol. 27 Iss 2 pp. 101-123 http://dx.doi.org/10.1108/ARJ -06-2012-0046
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Remuneration committee
independence and CEO
remuneration for ?rm
?nancial performance
Patti Cybinski
Department of Accounting, Finance & Economics, Grif?th University Nathan,
Brisbane, Australia, and
Carolyn Windsor
Faculty of Business, Bond University, Gold Coast, Australia
Abstract
Purpose – As a result of the Australian Government Productivity Commission’s recommendation to
mandate remuneration committee independence for ASX300 companies, this study aims to investigate
whether voluntary remuneration committee independence aligns chief executive of?cer (CEO) total
pay and bonuses with ?rm ?nancial performance.
Design/methodology/approach – A series of hypotheses test the research question using multiple
regressions for a sample of 143 ASX300 companies during 2001. This time was prior to strengthen
corporate governance regulation, but after mandated executive remuneration disclosure, thus
capturing varying levels of voluntary remuneration committee independence.
Findings – This study shows ?rm size is an in?uential factor in the relationship under investigation.
ASX300 large ?rm remuneration committees link CEO total remuneration and bonuses to ?rm
?nancial performance. Smaller ASX ?rm remuneration committees do not link either type of CEO
remuneration to performance despite remuneration committee independence. Findings are mixed for
medium-sized ASX300 ?rms.
Research limitations/implications – Limitations include the necessary time restriction to 2001 for
sampling the ASX300 ?rms. The implication of this study’s ?ndings is that the proposed public policy
for mandatory remuneration committee independence is not universally effective in linking CEO
remuneration to ?rm ?nancial performance for ASX300 ?rms.
Originality/value – This study contributes to the limited research on voluntary remuneration
committee independence in relation to CEO remuneration and ?rm ?nancial performance in the
Australian context.
Keywords Chief executive of?cer, Firm ?nancial performance, Remuneration committee independence
Paper type Research paper
1. Introduction
Public anger about executive pay excesses has increased since corporate, regulatory,
and political leaders were implicated in the global ?nancial crisis that harmed
community well-being (Stiglitz, 2010). Although Australian executive pay is relatively
modest by international standards, compensation for Australian executives in the
50 to 100 largest companies has increased by as much as 300 per cent in real terms
between 1993 and 2007 (Fels, 2010). Moreover, nearly all the growth in reported
executive pay for the top 300 companies is attributed to increases in incentive pay
(Fels, 2010).
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
Accounting Research Journal
Vol. 26 No. 3, 2013
pp. 197-221
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/ARJ-08-2012-0068
Remuneration
committee
independence
197
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Australian governments however have been successfully persuaded to let the market
regulate corporate executive remuneration. Statements of good practice, such as the
ASXCorporate Governance Council’s (2003, 2007) Corporate Governance Principles and
Recommendations “Remunerate fairly and responsibly”, largely guide remuneration
practice for Australian executives. Further, government legislative intervention is most
prevalent for remuneration disclosure and shareholders’ binding[1] vote on
remuneration (Sheehan, 2009, 2012). To appease public anger, the Prime Minister
requested the Productivity Commission[2] to review the regulation of executive and
director remuneration for the following reason:
[. . .] the prime motivation for this inquiry is a widespread perception that executives have
been rewarded for failure or simply good luck. And certainly in some periods and for some
CEOs, pay outcomes appear inconsistent with a reasonably ef?cient executive labour market
(Australian Government Productivity Commission, 2009, p. xxvi).
Our study is motivated by the Commission’s recommendation to regulate the formation
and composition of independent remuneration committees for companies included in
the S&P ASX 300 index (hereafter referred to as ASX300 companies). The Commission
argues that independent remuneration committees reduce con?icts of interest. These
con?icts include executive board members who are able to make decisions about their
own pay (Australian Government Productivity Commission, 2009). This key
recommendation number (2) proposes a new ASX listing rule specifying that.
The ASX Corporate Governance Council should introduce an “if not, why not”
recommendation specifying that remuneration committees:
.
have at least three members;
.
comprise (NEDIRs), a majority of whom are independent;
.
be chaired by an independent director;
.
have a charter setting out procedures for non-committee members attending; and
.
meetings (Australian Government Productivity Commission, 2009, p. 9, xxxvii).
The research question under investigation is whether remuneration committees
composed entirely of independent directors effectively ensure that chief executive of?cer
(CEO) pay relates directly to ?rm ?nancial performance. Australian remuneration
committee composition is mainly voluntary and guided by the ASX Corporate
Governance Council (2007) best practice that formally recommends boards to establish
remuneration committees. The Council suggests that remuneration committees comprise
a majority of independent directors, chaired by an independent director with a minimum
of three members (Australian Government Productivity Commission, 2009). In 2011,
regulation of executive compensation standards has increased with the ASXintroducing
the listing requirement (ASX Listing Rule 12.8) that ASX 300 companies form a
remuneration committee comprising entirely of NEDIRs (Kent et al., 2012).
As a consequence, of the Commission’s key recommendationand the 2011 ASXlisting
rule the purpose of this study is to speci?cally examine whether independent
remuneration committees reduce con?icts of interest by effectively linking CEO pay
with ?rm?nancial performance for ASX300 companies in 2001. Since 2001 was a time of
unregulated formation and composition of remuneration committees, the regulatory
context at the time is central to this study in a similar way to Rainsbury et al. (2009)
ARJ
26,3
198
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
who examined the impact of voluntary formation of another board subcommittee, the
audit committee, in an unregulated NewZealand context. We examine the period 2001 to
capture whether ASX300 ?rm boards had voluntarily established remuneration
committees and to measure the impact of varying degrees of remuneration committee
independence. The year 2001 was just before major regulatory changes to strengthen
corporate governance, but after mandated executive remuneration disclosure in the
Australian context.
We examine this research question through a series of regression models that are
tested against their corresponding implied hypotheses about existing associations. This
study starts with the simplest association between CEOremuneration and ?rm?nancial
performance, then evolves to models that test the importance of other intervening
variables, which could impact on this association either individually or through their
interaction effects. The ?rst hypothesis examines whether a positive relationship exists
between CEO remuneration and ?rm ?nancial performance, since shareholders use
rewards to induce CEO agents to work on the owners’ behalf and reduce agency
opportunism (Fama, 1980; Fama and Jensen, 1983). A rejection of this hypothesis
indicates no evidence of CEOs being rewarded for their performance or, alternatively,
that the model is incomplete and that important intervening factors associated with the
CEO remuneration for ?rm ?nancial performance relationship have been overlooked.
Through further modelling, our study examines remuneration committee independence
and ?rm size. Interactions provide the opportunity to gain more insight about the
research question by analysing further than the initial result (Hayes and Matthes, 2009).
Since we expect that ?rm ?nancial performance does not act alone in relation to
CEO remuneration, we also include interaction effects that some of these factors may
have with the performance measure on remuneration. A two-way interaction between
?rm ?nancial performance and remuneration committee independence was ?rst
included in the model as a means of testing whether remuneration committee
independence aligns CEO remuneration with ?nancial performance. Finally, we
investigate the model with a three-way interaction to test the combined impact of ?rm
?nancial performance, remuneration committee independence, and ?rm size on CEO
remuneration since all of these factors could in?uence CEO remuneration. We then
repeat the modelling process for CEO bonuses.
This study contributes to the limited research on voluntary remuneration committee
independence, in relation to CEO remuneration and ?rm ?nancial performance, prior to
the proposed regulation of remuneration committees. Further, this study’s ?ndings
contribute towards more informed public policy associated with corporate governance
about the impact of regulation on remuneration committee formation and composition
since the Productivity Commission’s (2009) recommendation to mandate remuneration
committee independence for ASX300 companies. The Commission made its
recommendation on the basis that these ASX300 ?rms are considered large and large
?rms have the resources to establish effective remuneration committees to ensure CEOs
are paid for performance (Australian Government Productivity Commission, 2009).
We ?nd that ?rm size is signi?cantly associated with the link between CEO
remuneration, remuneration committee independence, and ?rm ?nancial performance.
The study shows, however, that ASX300 ?rms are not homogeneous in relation to
remuneration committee independence aligning CEO remuneration to ?nancial
performance. ASX300 large ?rm remuneration committees align CEO total
Remuneration
committee
independence
199
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration and bonuses to ?rm ?nancial performance. Despite their remuneration
committee independence, smaller ASX300 ?rm remuneration committees are weakly
linked to CEO remuneration including bonuses and ?nancial performance. Findings
are mixed for ASX300 medium-size ?rms with the most independent remuneration
committees aligning CEO bonuses with performance.
The remainder of the paper is organised in the following way: Section 2 discusses
studies relevant to the dependent and independent variables in the model development
process; while Section 3 explains the research design. Section 4 tests the research
question through a series of regression models and explains the analysis and Section 5
concludes the study.
2. Variables and relevant studies
2.1 Dependent variables: CEO total remuneration and bonuses
The dependent variables in this study are ASX300 Australian companies’ CEO total
remuneration and CEO bonuses in dollar terms. CEO bonuses, in particular, have
received public attention because cash incentives in the form of bonuses have grown
exponentially with more emphasis on performance-related pay for company executives
since the 1990s (Australian Government Productivity Commission, 2009). In this study,
CEOtotal remunerationcomprises the sumindollars of the ?xedcomponent (base salary,
fringe bene?ts, and superannuation) and a short-term incentive component (bonuses).
For example, the total remuneration of the Brambles Brambles, (2001) CEO comprised
salary, cashallowances, bonuses, superannuation, motor vehicle, andretirement bene?ts.
Base salary is a ?xed form of remuneration and is normally contingent on the
leadership skills and experience of CEOs. Similarly, fringe bene?ts such as motor
vehicles and superannuation payments are not contingent on performance criteria,
per se, and their relative amounts vary depending on the remuneration package
negotiated with the employer company. Jensen et al. (2004) indicate that these ?xed
forms of remuneration should not be excessive when compared to the size or
performance of the ?rm. Also, any increases in ?xed remuneration should be partly
contingent on improved ?rm ?nancial performance.
Share options are not included in this study for the following reasons, although share
options are an important element of CEO remuneration ( Jensen et al., 2004). First, CEO
total remuneration is expressed in dollar terms, which makes it dif?cult to include share
options because of incomplete and vague reporting about their dollar value. Dollar value
also varies depending on when share options have been, or will be, exercised. If share
options are not traded there is no observed market valuation that can be included in
executive pay (Hutchinson and Gul, 2004). Moreover, the features of managerial share
options are such that they have little resemblance to options on stocks or securities in
general. This also makes it dif?cult to accurately value themusing the Black andScholes
(1973) option valuation model. Second, target-based incentive plans were by far the most
widely used for Australian CEO remuneration in our 2001 sample companies
(Hay Group, 1998). Cash (rather than shares or share options) is the predominant type of
payment under these plans with the requirement of being employed at the beginning of
the period (Godfrey et al., 2003). Thus, two-thirds of our sample 2001 companies either
did not compensate or report their CEO share options.
Jensen et al. (2004) argued that there is no optimal remuneration solution as
non-economic organisational features such as culture, structure, and strategies must
ARJ
26,3
200
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
be considered. Executive remuneration is negotiated in a largely self-regulated labour
market for senior company executives, resulting in a variety of executive remuneration
packages to enhance ?exibility to suit each company’s characteristics and goals
(Sheehan, 2009). Therefore, our ?nal sample of companies varied considerably in their
remuneration structure. Not all the companies disclosed remuneration elements (such as
superannuation), while other companies did not remunerate their directors with fringe
bene?ts (such as motor vehicles). Because base salary, fringe bene?ts, and
superannuation are ?xed forms of remuneration, it was possible to combine the three
forms of compensation into a total remuneration package. For this study, we examined
CEO total remuneration and CEO bonuses as dependent variables.
2.2 Independent variables
In the following section, we discuss the relevant research associated with the
independent variables that are linked to CEO remuneration and bonuses including ?rm
?nancial performance, remuneration committee independence, and ?rm size.
2.2.1 Firm ?nancial performance measure. The UK Cadbury Committee report
(Cadbury, 1992) was one of the earliest investigations into the integrity of corporate
governance. This committee recommended a voluntary code of best practice to include
a remuneration committee, as described by Ezzamel and Watson (1998, p. 222):
The remuneration of executive directors (those who are executives of the company on whose
board they sit) should be subject to the recommendations of a remuneration committee
(made up wholly or mainly of non-executive directors), which should ensure that large pay
awards were justi?ed by increases in ?rm performance and shareholders’ wealth.
Corporate boards argue that senior executives, particularly the CEO, should be
rewarded for managing a pro?table company that provides appropriate returns to the
owners, and, at the same time, for reducing the agency problem or management
opportunism (Fama and Jensen, 1983; Fama, 1980). From a managerialism perspective,
Berle and Means (1932, p. 25) identi?ed ?aws in the separation of ownership and
control, stating that “the separation of ownership from control produces a condition
where the interests of owner and of ultimate manager may, and often do, diverge, and
where many of the checks which formerly operated to limit the use of power disappear”
(cited Tosi et al., 2000, p. 302). Managerialist logic argues that without external
constraints, executives are more focused on increasing the ?rm size to increase their
pay rather than increasing pro?ts and ?rm performance (Tosi et al., 2000).
The separation of ownership from control inspired agency theory ( Jensen and
Meckling, 1976), which is based on the premise that principals (the shareholders)
delegate duties to an agent (the CEO), who is expected to act in the interest of the
principal. Agency theory is widely applied in corporate governance research because it
simpli?es assumptions about the two participating parties, shareholders and managers
whose interests are assumed to be clear and consistent (Daily et al., 2003). Agency theory
makes assumptions about managers’ behaviour: to illustrate, agents are assumed to
be risk averse and self-centred and therefore, agents’ interests may differ from those of
the principal. Opportunistic behaviour by the agent is possible as the agent may have
different objectives from the principal and, thus, pursue a self-serving agenda. The
board, through the remuneration committee, use a variety of incentives such as bonuses
and options in an attempt to align management pay with ?rm ?nancial performance in
the interest of shareholders ( Jensen, 2000; Jensen et al., 2004).
Remuneration
committee
independence
201
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Evidence is mixed regarding the link between CEO remuneration and ?rm
?nancial performance. While Merhebi et al. (2006) found a positive relationship
between CEO remuneration and ?rm ?nancial performance in their Australian study,
earlier Australian research for the years 1987-1992, by Izan et al. (1998), reported
no link between CEO remuneration and ?rm ?nancial performance. A later study
that examined Australian board structural independence (1999-2006) reported that
independent boards were no better than executive dominated boards in enforcing CEO
remuneration for ?nancial performance (Capezio et al., 2011).
Shaw and Zhang (2010) argued, however, that remuneration committees are reluctant
topunishCEOremunerationfor poor earnings, whenthose poor earnings result, inpart, on
investment decisions that are expected to generate positive future returns. Examples of
positive investment decisions are research and development expenditures or
restructurings. Using a yearly change in return on assets (ROA) as one proxy for
performance, Shaw and Zhang (2010) found in their US study that punishment in CEO
cash compensation for poor ?rm ?nancial performance does not systematically occur.
Their research also examined the CEO remuneration-performance relation for ?rms with
strong corporate governance. Their results provided further doubt on the role of ex post
settlingupinCEOcashcompensation. Inaddition, theyinvestigatedthe role of accounting
earnings in CEO compensation plans. Shaw and Zhang (2010) found that CEO
remuneration-performance sensitivity weakens for poor earnings performance, which is
consistent with studies that ?nd CEO cash compensation is shielded from transitory
accountinglosses (Gaver andGaver, 1998) andrestructuringcharges (Dechowet al., 1994).
ROA is, therefore, the primary measure of company ?nancial performance in this
study because companies typically base the incentive components of CEOremuneration
on ?nancial accounting based measures (Merhebi et al., 2006; Shaw and Zhang, 2010).
Further, the Australian Productivity Commission (2009, p. 70) stated that:
Researchers investigating the relationship between pay and corporate performance are
limited to using publicly-available indicators. Some have used regression analysis to identify
statistical relationships between indicators of corporate performance and executive
remuneration. This approach is reasonable provided the data are adequate and the
indicators correspond with those used by companies to determine executive remuneration.
In this study, we lag ROA by one year as we expect CEO compensation to be based on
?rm ?nancial performance for the prior year.
2.2.2 Remuneration committee independence. Some executives sit on boards, thus
creating con?icts of interest when setting executive pay. To reduce con?icts of interest
the Cadbury Committee (1992) recommended that the remuneration committee comprise
a majority of NEDIRs to enhance independent decision-making about executive pay.
Remuneration committees generally consist of independent NEDIRs to form an arm’s
length group to enhance the integrity of the decision-making process. The Corporations
Act 2001 (Cth) does not de?ne independence of directors, and Australian boards
generally adopt their own de?nition. The ASX Corporate Governance Council (2007)
de?nes an independent NEDIRs as one “who is free of any business or other
relationship that could materially interfere with the independent exercise of their
judgment, or be perceived to do so” (as cited in Australian Government Productivity
Commission, 2009, p. 174).
In 1993, independent remuneration committee composition became core US
regulation introduced by the Securities and Exchange Commission (SEC) and the
ARJ
26,3
202
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Internal Revenue Service (Vafeas, 2003). These regulations were motivated by the
concern that when executive directors participated in remuneration committees’
executive pay contracts, the remuneration committee was compromised resulting in pay
contracts that favoured management (Vafeas, 2003).
Vafeas (2003) investigated the impact of the US regulations on the requirement of
remuneration committees to comprise a majority of NEDIRs. After analysing 271 ?rms
for the 1991-1997 period, he reported a declining trend in executive director
membership of remuneration committees However, based on the evidence, he could not
conclude de?nitively that shareholders bene?ted from the US regulations to enhance
remuneration committee independence. Vafeas (2003) suggested that ?rms with
remuneration committees comprising outsiders were able to fend off uninvited public
scrutiny of pay practices and complying with the SEC’s 1992 disclosure rules about the
governance of executive compensation.
Conyon and Peck (1998), in their UK study of remuneration committees and executive
pay for 94 companies in the period 1991-1994, reported the positive relation between the
proportion of NEDIRs on a remuneration committee and senior management pay and
sensitivity of pay to performance. In contrast, Daily et al. (1998) found no link between
excessive CEO remuneration and remuneration committee dominance by executive
directors in their study of 194 US?rms for the period of 1991. These ?ndings are supported
by Newman and Mozes (1999), who analysed 1992 US company data for 161 ?rms and
found no relationship between CEO remuneration and executive director participation in
the remuneration committee. However, they noted that under certain conditions executive
pay for performance is skewed in management’s favour.
Sun et al. (2009) argued that because the SEC rules mandated wholly independent
compensation committees for US listed companies in 2003, a new and
more comprehensive measure of compensation committee quality was required. To
measure compensation committee quality, in relation to stock option grants and future
?rm performance, their study introduced and factor-analysed the following six
characteristics: CEO appointed directors, long-serving outside directors, CEOs from
other ?rms, high director stock ownership, busy outside directors (with three or more
additional board seats), and larger boards.
This study concentrates on remuneration committee independence, as the Australian
Government Productivity Commission (2009, p. xxxvii) proposed public policy to
mandate remuneration committee independence stated in their Recommendation 2 for
ASX300 ?rms. The aim of this study, therefore, is to investigate whether remuneration
committees comprising NEDIRs link CEO remuneration and bonuses with performance
of Australian companies for 2001. In our study, the proportion of NEDIRs comprising the
remuneration committee measures independence, de?ned by the ASX Corporate
Governance Council (2003, 2007), and is categorised into three levels for the regression
analyses that included factor interactions (Table I and II).
2001 company reports ASX300
Total ?rms in the dataset 300
Less trusts and unavailable ?rms 2001 reports due to
insolvencies, mergers, or no longer trading on ASX
(124)
Less ?rms with incomplete data (33)
Total sample 143
Table I.
Remuneration
committee
independence
203
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2.3 Firm size and executive compensation. Firm size does in?uence executive pay, as
larger companies seem to be prepared to pay for increased job importance and
complexity to attract more talented people, thus explaining 25-50 per cent observed
increases in executive pay in larger ?rms (Australian Government Productivity
Commission, 2009). Tosi et al. (2000) examined the links between ?rm size, ?rm
performance, and CEO remuneration in their meta-analysis, which formed part of
a review of the empirical literature on the determinants of CEO remuneration. Their
hypotheses speci?cally focused on:
.
?rm performance and CEO remuneration; and
.
?rm size and CEO remuneration.
They found that ?rm size accounted for more than 40 per cent of the variance in total
CEO remuneration, while ?rm performance accounted for less than 5 per cent of the
variance. Merhebi et al. (2006) tested the link between ?rmsize and CEOremuneration of
Australian companies (1990-1999) and found a signi?cant and positive result. They
argued that some evidence exists regarding size as a proxy for performance and that
“larger ?rms clearly have the capacity for higher remuneration packages regardless of
performance” (Merhebi et al., 2006, p. 495). Baker et al. (1988) questioned the notion of
size as a proxy for performance when they found evidence that CEOs are able to increase
their pay by increasing the ?rm size, even when the size increase reduces the ?rm’s
market value. They further suggested this motivation “could explain some of the vast
amount of inef?cient expenditures of corporate resources on diversi?cation programs
that have created large conglomerate organizations” (Baker et al., 1988, p. 609).
Additionally, Tosi et al. (2000, p. 329) argued that their ?ndings were consistent with:
[. . .] those theoretical explanations that emphasize organizational size as an important
determinant of total CEO remuneration; that is, indicators of ?rm size, taken together, explain
almost nine times the amount of variance in total CEO remuneration than the most highly
correlated performance measure. A lesser effect is demonstrated in the ?ndings regarding
pay sensitivity as well as in the difference in the pay/performance or pay/?rm growth
sensitivities. Changes in ?rm performance account for only 4 per cent of the variance in CEO
remuneration, while changes in ?rm size account for 5 per cent of the variance in CEO
remuneration. These results are consistent with Jensen and Murphy’s (1990) conclusion that
“incentive alignment” as an explanatory agency construct for CEO remuneration is weakly
supported at best.
Anumber of measures of ?rmsize appear in the literature, including sales revenue, log of
net sales, net income, total assets (Hagerman and Zmijewski, 1979; Skinner, 1994), and
log of total assets (Reynolds et al., 2004). Hagerman and Zmijewski (1979) argue that no
measure of size is necessarily better than another. Therefore, we use total assets as a
proxy for ?rm size and partition our sample into thirds for large, medium, and smaller
?rms to normalise the data (see Table III for mean total assets for each category).
3. Research design
Sample
To capture the varying levels of voluntary remuneration committee independence, we
expressly analyse data from ASX300 companies’ 2001 annual reports. The timeframe,
2001, is of interest because it is just after mandated executive remuneration disclosure,
but before the introduction of regulation to strengthen corporate governance including
ARJ
26,3
204
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ASX recommendation for remuneration committee independence. In fact, little
variation now exists in remuneration committee composition of ASX300 companies as
most now have independent remuneration committees. Further, Clarkson et al. (2011)
concluded that extensive regulatory change between 2001 and 2009 had enhanced
oversight of the executive compensation process through strengthened corporate
governance regulations and increased transparency.
Overview of increased corporate governance regulation post-2001
We, therefore, examine ASX300 companies in a corporate governance context before the
enactment of legislation that increased regulatory scrutiny of directors’ responsibilities
in response to corporate collapses, such as Enron in December 2001. The US Congress
enhanced corporate regulation by enacting the Sarbanes-Oxley Act (SOX) (2002). These
new corporate governance rules and requirements affected US-based trading ?rms and
Australian companies associated with the US capital markets. Additionally, Australian
based transnational audit ?rms began to use elements of SOX in their audits of
Australian companies. The Australian Government’s Company Law Economic Reform
Program (CLERP), concerning corporate governance, soon followed and was enacted
as CLERP 9 in 2004. Moreover, the ASX Corporate Governance Council’s (2003, 2007)
Corporate Governance Principles and Recommendations were introduced and
encouraged ?rms to implement independent remuneration committees that included a
majority of NEDIRs.
Mandated executive remuneration disclosure pre-2001
Australian regulation mandated disclosure of directors’ and executives’ remuneration
with the introduction of the Company Law Review Act 1998, revised May 1999
(Australian Securities and Investments Commission, 1998). Section 300A required
executive remuneration to be disclosed by listed Australian companies for ?nancial
years after 1 July 1998. The Act requires ?nancial statement disclosure of emoluments
of the ?ve most highly remunerated of?cers. Originally, executive remuneration was
disclosed in bands and not the actual amounts paid to executives. This requirement
was removed and the accounting standard required companies to disclose aggregate
remuneration of all executive of?cers whose remuneration for the ?nancial year is
$100,000 (Australian Accounting Standards Board (AASB) (1999) 1034).
The amended AASB 1034 became operative for companies at ?nancial year end
30 June 2001, so our sample of companies covers this speci?c period for the end of
the ?nancial year 2001. AASB (2005) 101 superseded this standard in July 2004
Level n
REMCindep
0 Lowest proportion: ?rms with no remuneration
committee or a minority of NEDIRs – least
independent
27
1 Firms with a majority of NEDIRs on remuneration
committees
42
2 100 percent NEDIRs on remuneration committees –
most independent
74
Total 143
Table II.
Remuneration committee
independence as
proportion of NEDIR at
three levels
Remuneration
committee
independence
205
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(operative 1 January 2005) to comply with the International Accounting Standards and
executive remuneration was speci?cally guided by AASB (2004) 1046 Director and
Executive Disclosures by Disclosing Entities.
Events in?uencing ?nal sample
Other contextual events in?uenced the collection of our sample of 2001 company
?nancial reports. These events include the internet bubble crash in March 2001 that
resulted in several corporate collapses and general investor turmoil in the following
years’ ?nancial reports (Boyer, 2005). For example, we looked for several IT companies
that were in our 2001 sample, but found they were non-existent in 2002, either insolvent
or merged. The 11 September 2001 attack on New York’s Twin Towers also resulted in
further investor turmoil that exacerbated the impact of the internet bubble crash.
Additionally, we found ASX300 company composition after 2001 was affected by high
changeability (survivorship) due to mergers, insolvencies, and, to some extent, the
re-privatising of public companies, making it dif?cult to sustain a longitudinal study
with a reasonable sample of the original 2001 ASX300 companies.
Our ?nal sample of companies in this study consists of 143 large publicly listed
Australian companies. The companies were chosen from the S&P/ASX 300[3] database
for 2001. Given that compulsory disclosure of corporate governance practices and
executive remuneration has only existed since late 1998 in Australia, extending the scope
of this study prior to this period would be pointless, as appropriate data to carry out the
study would not be publicly available. Further, we concluded that the effectiveness of
various committees, suchas remunerationcommittees, couldnot be evaluatedbefore 2000
because these committees were in the process of being established in many companies.
We individually downloaded S&P/ASX300 company reports for 2001 from
FinAnalysis, supplemented with data from Connect4 databases. From the 300 ?rms,
the sample consisted of 176 ?rms after excluding 124 ?rms that comprised trusts and
?rms whose reports were unavailable owing to insolvencies, mergers or no longer
trading. Out of the initial 176 companies, a ?nal sample of 143 companies had the
appropriate CEOremuneration and remuneration committee information. Surprisingly,
our sample included 45 companies that chose not to compensate their CEOs with
incentives. This is contrary to the recommendations of various committees, such as the
Cadbury Committee, and previous research (Baber et al., 1996; Beatty and Zajac, 1994;
Brunello et al., 2001; Carpenter and Sanders, 2002) which suggests that an effective
mechanism for aligning management’s interest with that of shareholders is to
compensate managers with short- and long-term incentives that are linked to ?rm
?nancial performance. Table I describes the sample used in this study.
Each report in our ?nal sample of 143 was manually examined to collect information
pertinent to this study. This hand-collected data includedthe dollar amounts for the CEO
remuneration components required for the dependent variables. The proportion of
NEDIRs on the remuneration committee was also hand collected for the categorical
independent variable remuneration committee independence. Essentially, the sample
companies had to disclose base salary, superannuation, fringe bene?ts, bonuses, share
options, and corporate governance information about the composition of remuneration
committees. FinAnalysis provided data for accounting information, ROA, and dollar
value of total assets to measure the independent variables, ?rm ?nancial performance,
and ?rm size.
ARJ
26,3
206
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.1 Measurement of variables
CEOtotal is the dependent variable measured as the natural logarithm of CEO
remuneration comprising the dollar sum of base pay, superannuation, fringe bene?ts,
and bonuses reported in 2001 ASX300 company reports.
CEObonus is the other dependent variable measured as the natural logarithm of the
short-term incentive for performance (in dollars) as part of CEO total remuneration
from the 2001 ASX300 company reports.
ROAlag1 is a continuous independent variable that is a measure of company
performance resulting from management’s productive use of company assets. ROA is
calculated as earnings before interest (total assets less outside equity interests) and is a
key measure of a company’s pro?tability, equal to a ?scal year’s earnings divided by its
total assets. ROAessentially shows howmuch pro?t a company is making on the assets
used in its business. ROA is lagged, as CEO compensation in our 2001 sample would
have been based on ?rm ?nancial performance for the prior year (2000).
REMCindep is measured as a proportion (percentage) of NEDIRs in the committee.
It was calculated by summing the number of NEDIRs in the committee and dividing
that ?gure by the total number of directors in that committee. Table II reports the
categorisation of REMCindep into three levels as follows: companies with no
remuneration committees or remuneration committees with a minority proportion of
(NEDIRs) are designated as 0; companies with a majority proportion less than 100 per
cent of NEDIRs on the remuneration committee are designated as 1; and, those
companies having all NEDIRs on the remuneration committee are designated as 2.
Size is an independent variable measured as total assets ($ value) categorised into
three levels of approximately equal number of ?rms: smaller ?rms ¼ 0, medium
?rms ¼ 1, and larger ?rms ¼ 2 (with the mean total assets for each category shown in
Table III). Transformation of ?rm Size into log total assets is unnecessary as this is
a categorical variable and, therefore, treated as a categorical variable in the
regression models.
Industry is a categorical variable to control for industry differences. Industry groups
comprise the ten sectors of the Global Industry Classi?cation Standard (GICS) used by
the ASX for industry classi?cation (Australian Securities Exchange, 2010a, b). For
analysis purposes, this study classi?es ten industry sectors into four industry groups
based on similarities in the nature of the industry. As outlined in Table IV, Panels Aand
B, industry is categorised: energy and mining ¼ 1, industrials ¼ 2, consumer ¼ 3, and
services ¼ 4.
4. Model development and results
Through a series of evolving models, we test the research question: whether
remuneration committee independence (REMCindep) effectively links CEO total
Size at three levels Level Mean of assets $M SD n
Smaller ?rms 0 148 779 46
Medium-sized ?rms 1 787 37 49
Larger ?rms 2 18,980 47,950 48
Total mean 143
Table III.
Firm size (total asset –
$ millions) at three levels
Remuneration
committee
independence
207
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration (CEOtotal ) and bonuses (CEObonus) with prior year’s ?rm ?nancial
performance (ROAlag1) for the period of 2001.
A higher ROA indicates management’s ability to use company assets ef?ciently in
serving shareholders’ economic interests. Therefore, we expect a positive association.
The link between CEO total remuneration and ?rm ?nancial performance is ?rst tested
without taking into consideration the effect of remuneration committee independence.
We use the general linear model (GLM) procedure in SPSS. This is a ?exible statistical
model that incorporates normally distributed dependent variables and categorical or
continuous independent variables (McCullagh and Nelder, 1989; Nelder and
Wedderburn, 1972). GLM allows factors (and interactions of factors) as predictors,
rather than just continuous variable predictors, in the models we are estimating; and is
appropriate because this research relied on considering the interactions between factor
variables.
Model 1: regression – CEO total remuneration and ?rm ?nancial performance
CEOtotal ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
Model 1 showed a non-signi?cant relationship for the link between CEO
remuneration and ?rm ?nancial performance (Table V, p ¼ 0.977). We argue,
however, that Model 1 is incomplete and other important factors are involved in
explaining CEO remuneration outcomes. Other factors deemed important in the
literature (i.e. remuneration committee independence and ?rm size) are tested by
considering Model 2.
Model 2 examines and tests the impact of an independent remuneration committee
as a corporate governance intervention. The independence of the remuneration
committee is represented by the proportion of NEDIRs on that committee, categorised
into three levels as follows: 74 ?rms had remuneration committees comprising
100 per cent of NEDIRs (i.e. the most independent remuneration committees); 42 ?rms
had remuneration committees with majority NEDIR composition; and 27 ?rms had no
Panel A: industry groups based on GICS code and industry sector
Industry groups GICS code Sector
1. Energy and mining 10 Energy
15 Materials
2. Industrials 20 Industrials
3. Consumer 25 Consumer discretionary
30 Consumer staples
4. Services 35 Health care
40 Financials
50 Telecommunication services
55 Utilities
Panel B: the no. of ?rms for each industry group
Industry sector
1. Energy and mining 33 (23.132%)
2. Industrials 21 (14.673%)
3. Consumer 44 (30.674%)
4. Services 45 (31.521%)
Table IV.
ARJ
26,3
208
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration committees or remuneration committees with a minority of NEDIRs
(i.e. the least independent remuneration committees, see Table II).
Model 2: regression, a two-way interaction model for CEO total remuneration on ?rm
?nancial performance and remuneration committee independence
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
where: b1 is now a vector of coef?cients for the various levels of the interaction term in
the model.
In Model 2, we test the effect of a two-way interaction between ?rm ?nancial
performance and remuneration committee independence on CEO remuneration. This is
to investigate whether remuneration committees control CEO pay in line with ?rm
?nancial performance in at least some partitions of the group of ?rms characterized by
the different levels of remuneration committee independence. For theoretical reasons,
the main effects ROAlag1 and REMCindep are not included because the interaction
effect between themis signi?cant in Model 2 ( p ¼ 0.01). To set up a model that includes
these two main effects, and interpret their coef?cients and hypothesis tests as main
effects, is not theoretically justi?ed or correct when their interaction effect is signi?cant,
except under limited conditions (Hayes, 2005, pp. 452-456; Irwin and McClelland, 2001;
Jaccard and Turrisi, 2003, p. 24).
The two-way interaction model for CEOtotal remuneration is signi?cant (F
3,136
¼ 3.7;
p ¼ 0.01, Table VI) indicating that independent remuneration committees are
signi?cantly associated with the relationship between CEO total remuneration and
prior year’s ?rm ?nancial performance. We examine the three coef?cient parameter
estimates for the interaction term in Table VII to ascertain in which subgroup/s of
remuneration committee independence the association between CEO remuneration and
?rm?nancial performance is signi?cant, and whether it is a positive or negative one. We
?nd that for the most independent remuneration committees (with 100 per cent NEDIR
composition) the association between performance and CEO remuneration is signi?cant
and positive ( p ¼ 0.06, Table VII). However, for the least independent remuneration
committees (with either no remuneration committee or a minority of NEDIRs) the
association between ?rm?nancial performance and CEOremuneration is signi?cant and
negative ( p , 0.05, Table VII).
Variable F p-value
Intercept 26,256.166 0.000
Industry 0.147 0.932
ROAlag1 0.004 0.952
Likelihood ratio x
2
¼ 0.462, df ¼ 4 ( p ¼ 0.977)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; industry – industry
groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table V.
Model 1 CEO total
remuneration and ?rm
?nancial performance
Remuneration
committee
independence
209
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This explains the lack of signi?cance for Model 1, since the CEO remuneration for ?rm
?nancial performance relationship is neither consistently signi?cant nor consistently
positive for all three levels of remuneration committee independence. The model ?t is
signi?cant ( p ¼ 0.07 , 0.1), using the maximum likelihood ratio x
2
test. However,
the model ?t can be improved by including ?rmsize as a three-way interaction termwith
the other two independent variables.
Variable b SE t p-value
Intercept 13.67 0.158 86.467 0.000
Energy and mining 20.056 0.210 20.265 0.792
Industrials 20.163 0.241 20.675 0.501
Consumer 20.125 0.193 20.648 0.518
Services 0
a
ROAlag1
*
[REMCindep ¼ 0] 20.024 0.012 21.996 0.048
* *
ROAlag1
*
[REMCindep ¼ 1] 0.026 0.018 1.449 0.150
ROAlag1
*
[REMCindep ¼ 2] 0.025 0.013 1.883 0.062
*
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithmof CEOtotal remuneration ($) comprisingbase pay þ superannuation/fringe
bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; REMCindep is the proportion of
NEDIRs onthe remuneration committee (RC), categorised into 0 – ?rms with noRCs or a minorityNEDIRs,
the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but less than 100 per cent; 2 – 100 per cent
NEDIR composition, the most independent RCs; industry – industry groups based on GIC sectors, 1 –
energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VII.
Model 2 CEO total
remuneration parameter
estimates for two-way
Interaction between ?rm
?nancial performance
and remuneration
committee independence
Variable F p-value
Intercept 2,624.072 0.000
Industry 0.216 0.885
ROAlag1
*
REMCindep 3.717 0.013
* * *
Likelihood ratio x
2
¼ 11.7, df ¼ 6 ( p ¼ 0.068)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithmof CEOtotal remuneration ($) comprisingbase pay þ superannuation/fringe
bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; REMCindep is the proportion of
(NEDIRs) ontheremunerationcommittee (RC), categorisedinto0 – ?rms withnoRCs or aminorityNEDIRs,
the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but less than 100 per cent; 2 – 100 per cent
NEDIR composition, the most independent RCs; industry – industry groups based on GIC sectors, 1 –
energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VI.
Model 2 CEO total
remuneration and
two-way interaction
between ?rm
performance and
remuneration committee
independence
ARJ
26,3
210
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Model 3: regression – a three-way interaction model for CEO total remuneration on
?rm ?nancial performance, remuneration committee independence and ?rm size
Prior research by Tosi et al. (2000) and Merhebi et al. (2006) examined ?rm size and
?nancial performance separatelyinrelationto CEOremuneration. These studies suggest
that large ?rms generously pay their CEOs and executives and size was found to be a
more important criterion than ?nancial performance alone. Recall that our study
contributes to corporate governance research and associated public policy by examining
whether varying levels of remuneration committee independence align CEO total
remuneration with ?rm ?nancial performance for ASX 300 companies in an essentially
voluntary context. Although we examine ASX300 ?rms, these ?rms vary in size and are
not homogeneous. Firmsize (Size) is categorisedinto the followingthree levels (described
in Table III): smaller ?rms (n ¼ 46, mean 148, SD77 in $m), medium?rms (n ¼ 49, mean
787, SD 373 in $m), and large ?rms (n ¼ 48, mean 18,980, SD 47,950 in $m). This
categorisation of size allows analysis of the model with a three-way interaction below:
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
where: b1 is a vector of coef?cients for the various levels of the interaction term in the
three-way interaction model (Friedrich, 1982; Jaccard and Turrisi, 2003). For Model 3,
Table VIII reports a signi?cant high-order relationship (F
12,130
¼ 8.94, p ¼ 0.001)
between the dependent variable, CEO total remuneration (CEOtotal ) and the three-way
interaction with ?rm size (Size), prior year’s ?rm ?nancial performance (ROAlag1), and
remuneration committee independence (REMCindep). Model 3 is a much better ?t than
Model 2, as indicated by the maximum likelihood x
2
test ( p , 0.001). Table IX reports
further analysis of this three-way interaction revealing that ?rmsize matters, with very
signi?cant positive associations for large ?rms ( p , 0.01) between CEOtotal and ?rm
?nancial performance (ROAlag1) for all three levels of remuneration committee
independence (REMCindep).
The results for the three-way interaction are particularly strong ( p , 0.001),
especially for large ?rms with remuneration committees comprising 100 per cent
Variable F p-value
Intercept 12,175.206 0.000
Industry 1.469 0.226
Size
*
ROAlag1
*
REMCindep 8.940 0.000
* * * *
Likelihood ratio x
2
¼ 69.353, df ¼ 12 ( p ¼ 0.000)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; Size – total assets ($ value) categorised into thirds: 0 – smaller ?rms, 1 –
medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000; REMCindep is the
proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms with no RCs or a
minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs but less than
100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; industry – industry
groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VIII.
Model 3 CEO total
remuneration and
three-way interaction
between ?rm size, ?rm
?nancial performance
and remuneration
committee independence
Remuneration
committee
independence
211
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
NEDIRs and for ?rms with a majority of NEDIRs ( p ¼ ,0.001). Results are weaker
( p , 0.005) for ?rms with no remuneration committees or a minority NEDIRs. For
medium and smaller size ?rms, the only signi?cant positive association between CEO
total pay and prior year’s ?rm ?nancial performance occurred for medium-sized ?rms
with remuneration committees comprising a majority of NEDIRs ( p , 0.05).
Interestingly, a signi?cant negative relationship between CEOtotal and prior year’s
?rm ?nancial performance was found for smaller ?rms with no remuneration
committees or remuneration committees with a minority of NEDIRs (the lowest level of
remuneration committee independence). This last result suggests that smaller ?rms
with the least independent remuneration committees recompense their CEOs even when
these ?rms are not performing ?nancially. The results also show the ineffectiveness of
remuneration committees to align CEO total pay with ?rm ?nancial performance for
smaller ?rms regardless of the composition of the remuneration committee.
4.1 CEO bonuses
CEO bonuses represent short-term incentives that should be based directly on ?rm
?nancial performance. Therefore, similar to previous arguments, an independent
remuneration committee should control CEOincentive pay or bonuses for performance.
The testing process of Models 1-3 is a similar process as for Models 4-6, but with
CEO bonuses as the dependent variable.
Variable b SE t p-value
Intercept 13.554 0.141 96.452 0.000
Energy and mining 20.312 0.186 21.680 0.095
Industrials 20.171 0.204 20.838 0.404
Consumer 20.324 0.170 21.908 0.059
Services 0
a
[Size ¼ 0 smaller]
*
[REMCindep ¼ 0]
*
ROAlag1 20.023 0.011 22.184 0.031
* *
[Size ¼ 0 smaller]
*
[REMCindep ¼ 1]
*
ROAlag1 20.016 0.021 20.751 0.454
[Size ¼ 0 smaller]
*
[REMCindep ¼ 2]
*
ROAlag1 0.003 0.013 0.257 0.798
[Size ¼ 1 med]
*
[REMCindep ¼ 0]
*
ROAlag1 20.014 0.030 20.467 0.641
[Size ¼ 1 med]
*
[REMCindep ¼ 1]
*
ROAlag1 0.059 0.025 2.335 0.021
* *
[Size ¼ 1 med]
*
[REMCindep ¼ 2]
*
ROAlag1 0.032 0.025 1.311 0.192
[Size ¼ 2 large]
*
[REMCindep ¼ 0]
*
ROAlag1 0.152 0.051 2.970 0.004
* * *
[Size ¼ 2 large]
*
[REMCindep ¼ 1]
*
ROAlag1 0.126 0.025 4.991 0.000
* * * *
[Size ¼ 2 large]
*
[REMCindep ¼ 2]
*
ROAlag1 0.139 0.023 6.153 0.000
* * * *
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; Size – total assets ($ value) categorised into thirds: 0 – smaller ?rms, 1 –
medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000; REMCindep is the
proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms with no RCs or a
minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs but less than 100
per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; industry – industry groups
based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table IX.
Model 3 CEO total
remuneration parameter
estimates for three-way
interaction between ?rm
size, remuneration
committee independence,
and ?rm ?nancial
performance
ARJ
26,3
212
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Model 4: regression – CEO bonuses and ?rm ?nancial performance
CEObonus ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
Model 4, with CEO bonuses (CEObonus) as the dependent variable, is not signi?cant
( p ¼ 0.22) indicating that our sample of ?rms showed no evidence of aligning CEO
bonus pay with prior year’s performance (Table X). As before, we argue that this model
is incomplete and other intervening variables are involved. Thus, we include
remuneration committee independence in Model 5 to improve the model ?t, as in
Models 1-3 that tested CEO total remuneration as the dependent variable.
Model 5: regression – a two-way interaction model for CEO bonuses on ?rm ?nancial
performance and remuneration committee independence
CEObonus ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
In Model 5 we test CEO bonuses and the two-way interaction between ?rm ?nancial
performance and remuneration committee independence. The result was not
signi?cant ( p ¼ 0.238), as shown in Table XI. The two-way interaction shows only
marginal signi?cance ( p ¼ 0.1). We conclude that our sample of ?rms, when
partitioned by the independence level of their remuneration committee, show no
evidence of aligning CEO bonuses with prior year’s ?rm ?nancial performance within
any subgroup of remuneration committee categorised at three levels of independence.
Model 6: regression model – a three-way interaction model for CEO bonus on ?rm
?nancial performance, remuneration committee independence and ?rm size
Next, in Model 6, we investigate the association between the dependent variable CEO
bonus, and the three-way interaction between ?rm size, prior year’s ?rm ?nancial
performance, and remuneration committee independence. Table XII Model 6 reports a
signi?cant three-way interaction (F
9,129
¼ 2.24, p , 0.05), but not as strong as for
Model 3 that shows CEO total remuneration ( p , 0.001). Just as for CEO total
remuneration, Table XIII reports further analysis of Model 6 interaction and shows a
strong positive relationship between CEO bonuses and prior year’s ?nancial
performance (ROAlag1) for large ?rms at all three levels of remuneration committee
independence (REMCindep). The relationship is strongest ( p , 0.002) for large ?rms
Variable F p-value
Intercept 102.202 0.000
Industry 1.653 0.180
ROAlag1 0.589 0.444
Likelihood ratio x
2
¼ 5.73, df ¼ 4 ( p ¼ 0.22)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; ROAlag1 – return on assets for the prior year 2000;
industry – industrygroups basedonGICsectors, 1 – energyandmining, 2 – industrials, 3 – consumer and
4 – services
Table X.
Model 4 CEO bonuses
with ?rm ?nancial
performance
Remuneration
committee
independence
213
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
with remuneration committees comprising 100 per cent NEDIRs. This result shows
that the largest ?rms with the most independent remuneration committees consistently
paid CEO bonuses in line with ?rm ?nancial performance, thus demonstrating that
CEO bonuses are somewhat controlled by independent remuneration committees for
this subgroup of ?rms.
We ?ndaweaker result for large ?rms withremunerationcommittees havinga majority
of NEDIRs or mid-level independence ( p , 0.05); while, those ?rms with no remuneration
committees or a minority of NEDIRs were weaker still ( p , 0.01). Table XIII also reports a
positive signi?cant result for medium-sized ?rms with 100 per cent NEDIRs ( p ¼ ,0.05).
These results suggest that large and medium ?rm groups with 100 per cent NEDIRs on
their remuneration committees are the most effective in controlling CEO bonus incentives
in line with prior year’s ?rm ?nancial performance. The results also show the
Variable F p-value
Intercept 40.354 0.000
Industry 2.560 0.058
Size
*
ROAlag1
*
REMCindep 2.244 0.023
* *
Likelihood ratio x
2
¼ 25.7, df ¼ 12 ( p ¼ 0.012)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; Size – total assets ($ value) categorised into thirds: 0 –
smaller ?rms, 1 – medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000;
REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms
with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs
but less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; Industry –
industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 –
services
Table XII.
Model 6 CEO bonuses
and three-way interaction
between ?rm size, ?rm
?nancial performance,
and remuneration
committee independence
Variable F p-value
Intercept 106.083 0.000
Industry 1.703 0.169
ROAlag1
*
REMCindep 2.727 0.101
Likelihood ratio x
2
¼ 8.0, df ¼ 6 ( p ¼ 0.238)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; ROAlag1 – return on assets for the prior year 2000;
REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms
with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but
less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; Industry –
industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 –
services
Table XI.
Model 5 CEO bonuses
and two-way interaction
between ?rm ?nancial
performance and
remuneration committee
independence
ARJ
26,3
214
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ineffectiveness of remuneration committees to align CEO bonuses with ?rm ?nancial
performance for the smaller ?rms regardless of the composition of the remuneration
committee.
5. Discussion and conclusion
The Productivity Commission’s key draft recommendation (Australian Government
Productivity Commission, 2009, p. xxxiii) proposed a new ASX listing rule specifying
that “all ASX300 companies have a remuneration committee of at least three members,
all of whom are non-executive directors, with the chair and a majority of members
being independent” to avoid con?icts of interest when paying executives. Our study
examines whether remuneration committee independence aligns CEO compensation
with performance for ASX300 ?rms during 2001, a period of voluntary corporate
governance.
We show that ?rm size is an important factor in determining whether voluntary
remuneration committees align CEO remuneration with ?rm ?nancial performance.
Companies having remuneration committees comprising all NEDIRs do not always
align CEO total pay or CEO bonuses with ?rm ?nancial performance. Our study ?nds
that large ?rm remuneration committees align both CEO total remuneration and CEO
bonuses with ?rm ?nancial performance at all levels of remuneration committee
independence; although, the most effective committees (i.e. with model coef?cients
Variable b SE t p-value
Intercept 8.019 1.025 7.820 0.000
Energy and mining 20.510 1.357 20.376 0.708
Industrials 22.613 1.490 21.754 0.082
Consumer 22.983 1.246 22.394 0.018
Services 0
a
[Size ¼ 0 smaller]
*
[REMCindep ¼ 0]
*
ROAlag1 20.016 0.078 20.202 0.840
[Size ¼ 0 smaller]
*
[REMCindep ¼ 1]
*
ROAlag1 0.067 0.154 0.434 0.665
[Size ¼ 0 smaller]
*
[REMCindep ¼ 2]
*
ROAlag1 0.022 0.092 0.243 0.808
[Size ¼ 1 med]
*
[REMCindep ¼ 0]
*
ROAlag1 0.042 0.220 0.190 0.849
[Size ¼ 1 med]
*
[REMCindep ¼ 1]
*
ROAlag1 0.187 0.184 1.016 0.311
[Size ¼ 1 med]
*
[REMCindep ¼ 2]
*
ROAlag1 0.423 0.180 2.345 0.021
* *
[Size ¼ 2 large]
*
[REMCindep ¼ 0]
*
ROAlag1 1.015 0.374 2.718 0.007
* * *
[Size ¼ 2 large]
*
[REMCindep ¼ 1]
*
ROAlag1 0.359 0.184 1.948 0.054
* *
[Size ¼ 2 large]
*
[REMCindep ¼ 2]
*
ROAlag1 0.523 0.166 3.158 0.002
* * *
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEObonus ¼ b0 þ b1 ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; Size – total assets ($ value) categorised into thirds: 0
– smaller ?rms, 1 – medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year
2000; REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0
– ?rms with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on
?rm RCs but less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs;
Industry – industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 –
consumer and 4 – services
Table XIII.
Model 6 CEO total
bonuses, parameter
estimates for three-way
interaction between ?rm
size remuneration
committee independence
and ?rm ?nancial
performance
Remuneration
committee
independence
215
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
having the smallest p-values) have 100 per cent NEDIR membership. The study also
shows that smaller ?rmremuneration committees are ineffective in reducing con?icts of
interest in relation to CEO total remuneration and bonuses for ?rm ?nancial
performance regardless of the level of their remuneration committee independence.
Results for medium-sized ?rms are mixed. Those medium-sized ?rms with
remuneration committees with a majority of NEDIRs link CEO total remuneration
with ?rm ?nancial performance, while medium-sized ?rms with remuneration
committees comprising 100 per cent NEDIRs link CEO bonuses with ?rm ?nancial
performance.
Our evidence suggests that the Australian Government Productivity Commission’s
(2009) recommendation to mandate independent remuneration committees to align
CEO remuneration with ?rm ?nancial performance may not be effective public policy,
particularly for medium and smaller ASX300 ?rms. To address corporate CEO
remuneration excesses, policy makers need to be aware of additional factors, such as
remuneration committee quality (Sun et al., 2009), because larger ?rms have the
resources to attract and employ higher quality human and intellectual resources
including quality independent directors (Australian Government Productivity
Commission, 2009). Moreover, the CEOs of large ?rms are more likely to
achieve appropriate ?rm ?nancial performance supported by extensive resources
(Merhebi et al., 2006).
Our study shows that CEOs of smaller ?rms receive full remuneration (including
bonuses), but that this remuneration has weak links to performance. This ?nding
suggests that the CEOs of smaller ?rms appear to attract less public and market scrutiny
than their larger counterparts and that smaller ?rms pay their CEOs regardless of
whether these ?rms perform or not. In the US regulatory context, where independent
compensation committees have been mandatory since 2003, Sun et al. (2009) found that
higher compensation committee quality, comprising a multi-dimensional measure, led to
a greater incentive alignment in executive compensation contracts in relation to future
?rm ?nancial performance. Future research could compare the 2001 sample of ASX 300
?rms with a sample of 2012 ASX 300 ?rms to see the regulatory impact of increased
remuneration committee independence. Additionally, the quality of remuneration
committee composition in the more regulated Australian context could also be examined.
While our study provides evidence about remuneration committee independence
interventions in ASX300 companies, there are limitations. Our sample of ?rms was
restricted to 2001, a time when it was possible to capture varying levels of remuneration
committee independence prior to increased corporate governance regulations. As with
most social science research, the models used to test the hypotheses are not complete and
we acknowledge that other important factors or interaction terms may not have been
included when investigating the relationship between CEO total pay and ?rm ?nancial
performance. Multicollinearity, however, becomes a problem if too many ?nancial
variables are included in the same model. We also acknowledge that this study is
unlikely to be completely free of endogeniety because “the potential for endogeniety
exists for virtually all studies involving accounting, ?nance and economics variables”
(Chenhall and Moers, 2007, p. 177). As they suggest, we have attempted to deal with
potential endogeniety by identifying the dependent variable and
independent/explanatory variables to be included in the models based on theory and
prior empirical evidence.
ARJ
26,3
216
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Notes
1. The Corporate Law Economic Reform Program (Audit Reform and Corporate Disclosure)
Act (2004) included s 250R(2) in the Corporations Act 2001 (Cth) as part of the legislative
reforms to the disclosure of director and executive remuneration. “Section 250R(2) requires
that a resolution that the remuneration report (attached to the directors’ report) be adopted
be put to shareholder vote at the annual general meeting (AGM) of listed companies. Section
250R(3) provides that this vote is to be non-binding. While a majority of voting shareholders
may vote against the proposed remuneration, the ultimate decision whether to accept the
report rests with the board” (Chapple and Christensen, 2005, p. 263). Shareholders’
non-binding vote on remuneration appears to have resulted in improved communication
between boards and shareholders and enhanced remuneration practices, despite initial
opposition by the local business community (Fels, 2010). The problem is that some boards
have ignored shareholders’ non-binding vote regarding executive remuneration (Fels, 2010).
In fact, the Australian Corporations Act was amended in 2011 to enact shareholders’ binding
vote or the Two Strike Rule. When there are two successive negative votes on board
remuneration reports at annual general meetings by 25 per cent or more of shareholders, the
company must formally respond by asking all board members except the CEO to stand for
re-election within 90 days. One-third of all board positions in Australian public companies
are elected every year (Sheehan, 2012).
2. The Productivity Commission undertook a public inquiry into the regulatory framework
around the remuneration of directors and executives of companies regulated under the
Corporations Act. The report was presented to the Australian Government on 19 December
2009 and released publicly on 4 January 2010. Speci?cally, the Commission was requested to
consider the following: the effectiveness of the existing framework for the oversight,
accountability and transparency of remuneration practices in Australia including; the role,
structure and content of remuneration disclosure and reporting; the scope of who should be
the subject of remuneration disclosure and; approving remuneration packages; the role of
boards and board committees in developing and approving; remuneration packages
(Australian Government Productivity Commission, 2009, p. v).
3. “The S&P/ASX 300 provides up to an additional 100 small cap stocks to the S&P/ASX200.
Index constituents are drawn fromeligible companies listed on the Australian Stock Exchange.
This index is designed to address investment managers’ needs to benchmark against a
portfolio characterized by suf?cient size and liquidity. [. . .]” (www.standardandpoors.com/
indices/sp-asx).
References
AASB (2004), Accounting Standard AASB 1046 Director and Executive Disclosures by Disclosing
Entities, Australian Accounting Standards Board, Melbourne.
AASB (2005), Accounting Standard AASB 101 Presentation of Financial Statements, Australian
Government Australian Accounting Standards Board, Melbourne.
AASB (1999), Accounting Standard AASB 1034 Financial Report Presentation and Disclosures,
Australian Accounting Standards Board, Australian Accounting Research Foundation,
Melbourne.
Australian Government Productivity Commission (2009), Executive Remuneration in Australia,
Report No. 49, Final Inquiry Report, Productivity Commission, Melbourne, available at:
www.pc.gov.au
Australian Securities Exchange (2007), Corporate Governance Principles and Recommendations with
2010 Amendments, 2nd ed., ASX Corporate Governance Council, Sydney, available at: www.
asxgroup.com.au/media/PDFs/cg_principles_recommendations_with_2010_amendments.pdf
Remuneration
committee
independence
217
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Australian Securities Exchange (2010a), Full List of Current GICS Sectors, available at: www.asx.
com.au/products/sector-indices.htm#consumer_discretionary_index
Australian Securities Exchange (2010b), “What is GICS?”, available at: www.asx.com.au/
products/gics.htm
Australian Securities and Investments Commission (1998), Practice Note 68: New Financial
Reporting and Procedural Requirements, Australian Securities and Investments
Commission, available at: www.asic.gov.au/asic/pd?ib.nsf/LookupByFileName/pn68.pdf/
$?le/pn68.pdf
Australian Stock Exchange (2003), Principles of Good Corporate Governance and Best
Practice Recommendations, ASX Corporate Governance Council, Sydney, March,
available at: www.asx.com.au/documents/about/principles-and-recommendations-march-
2003.pdf
Baber, W.R., Janakiraman, S.N. and Kang, S. (1996), “Investment opportunities and the
structure of executive compensation”, Journal of Accounting & Economics, Vol. 21 No. 3,
pp. 297-318.
Baker, G.P., Jensen, M.C. and Murphy, K.J. (1988), “Compensation and incentives: practice vs
theory”, The Journal of Finance, Vol. 43 No. 3, pp. 593-616.
Beatty, R.P. and Zajac, E.J. (1994), “Managerial incentives, monitoring, and risk bearing: a study
of executive compensation, ownership, and board structure in initial public offerings”,
Administrative Science Quarterly, Vol. 39 No. 2, pp. 313-335.
Berle, A.A. Jr and Means, C.G. (1932), The Modern Corporation and Private Property,
The Macmillan Company, New York, NY.
Black, F. and Scholes, M. (1973), “The pricing of options and corporate liabilities”, Journal of
Political Economy, Vol. 81 No. 3, pp. 637-654.
Boyer, R. (2005), “From shareholder value to CEO power: the paradox of the 1990s”, Competition
& Change, Vol. 9 No. 1, pp. 7-47.
Brambles (2001), Brambles Industries 2001 Annual Report, Brambles Industries PLC,
available at: www.brambles.com/BXB/Company/ShowPage.aspx/PDFs/3253-10000000/
2001AnnualReport156KB
Brunello, G., Graziano, C. and Parigi, B. (2001), “Executive compensation and ?rm
performance in Italy”, International Journal of Industrial Organization, Vol. 19 Nos 1/2,
pp. 133-161.
Cadbury, A. (1992), Report of the Committee on the Financial Aspects of Corporate Governance,
Gee, London.
Capezio, A., Shields, J. and O’Donnell, M. (2011), “Too good to be true: board structural
independence as a moderator of CEO pay-for-?rm-performance”, Journal of Management
Studies, Vol. 48 No. 3, pp. 487-513.
Carpenter, M.A. and Sanders, W.G. (2002), “Top management team compensation: the missing
link between CEO remuneration and ?rm performance?”, Strategic Management Journal,
Vol. 23 No. 4, pp. 367-375.
Chapple, E. and Christensen, B. (2005), “The non-binding vote on executive pay: a review of the
CLERP 9 reform”, Australian Journal of Corporate Law, Vol. 18, pp. 263-287.
Chenhall, R.H. and Moers, F. (2007), “The issue of endogeniety in theory-based,
quantitative management accounting research”, European Accounting Review, Vol. 16 No. 1,
pp. 173-195.
ARJ
26,3
218
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Clarkson, P., Walker, J.K. and Nicholls, S. (2011), “Disclosure, shareholder oversight
and pay-to-performance link”, Journal of Contemporary Accounting and Economics, Vol. 7,
pp. 47-64.
Conyon, M.J. and Peck, S.I. (1998), “Board control, remuneration committees, and
top management compensation”, Academy of Management Journal, Vol. 41 No. 2,
pp. 146-157.
Daily, C.M., Dalton, D.R. and Cannella, A.A. Jr (2003), “Corporate governance: decades of
dialogue and data”, Academy of Management Review, Vol. 28 No. 3, pp. 371-382.
Daily, C.M., Johnson, J.L., Ellstrand, A.E. and Dalton, D.R. (1998), “Compensation committee
composition as a determinant of CEO compensation”, Academy of Management Journal,
Vol. 41 No. 2, pp. 209-220.
Dechow, P.M., Huson, M.R. and Sloan, R.G. (1994), “The effect of restructuring charges on
executives’ cash compensation”, The Accounting Review, Vol. 69 No. 1, pp. 138-156.
Ezzamel, M. and Watson, R. (1998), “Market comparison earnings and the bidding-up of
executive cash compensation: evidence from the United Kingdom”, Academy of
Management Journal, Vol. 41 No. 2, pp. 221-231.
Fama, E.F. (1980), “Agency problems and the theory of the ?rm”, Journal of Political Economy,
Vol. 88 No. 2, pp. 288-307.
Fama, E.F. and Jensen, M.C. (1983), “Separation of ownership and control”, The Journal of Law
and Economics, Vol. 26 No. 2, pp. 301-325.
Fels, A. (2010), “Executive remuneration in Australia”, The Australian Accounting Review, Vol. 20
No. 1, pp. 76-82.
Friedrich, R.J. (1982), “In defense of multiplicative terms in multiple regression equations”,
American Journal of Political Science, Vol. 26 No. 4, pp. 797-833.
Gaver, J.J. and Gaver, K.M. (1998), “The relation between nonrecurring accounting
transactions and CEO cash compensation”, The Accounting Review, Vol. 73 No. 2,
pp. 235-253.
Godfrey, J., Mather, P. and Ramsay, A. (2003), “Earnings and impression management in
?nancial reports: the case of CEO changes”, Abacus, Vol. 39 No. 1, pp. 95-123.
Hagerman, R.L. and Zmijewski, M.E. (1979), “Some economic determinants of accounting
policy choice”, Journal of Accounting & Economics, Vol. 1 No. 2, pp. 141-161.
Hay Group (1998), Short-term Incentives Report 1998/99, Hay Group, Melbourne.
Hayes, A.F. (2005), Statistical Methods for Communication Science, Erlbaum, Mahwah, NJ.
Hayes, A.F. and Matthes, J. (2009), “Computational procedures for probing interactions in OLS
and logistic regression: SPSS and SAS implementations”, Behavior Research Methods,
Vol. 41 No. 3, pp. 924-936.
Hutchinson, M. andGul, F.A. (2004), “Investment opportunityset, corporate governancepractices and
?rm performance”, Journal of Corporate Finance, Vol. 10 No. 4, pp. 595-614.
Irwin, J.R. and McClelland, G.H. (2001), “Misleading heuristics and moderated multiple
regression models”, Journal of Marketing Research, Vol. 38 No. 1, pp. 100-109.
Izan, H.Y., Sidhu, B. and Taylor, S. (1998), “Does CEO remuneration re?ect performance?
Some Australian evidence”, Corporate Governance: An International Review, Vol. 6 No. 1,
pp. 39-47.
Jaccard, J. and Turrisi, R. (2003), Interaction Effects in Multiple Regression, 2nd ed., Quantitative
Applications in the Social Sciences No. 72, Sage, Thousand Oaks, CA.
Remuneration
committee
independence
219
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Jensen, M.C. (2000), A Theory of the Firm: Governance, Residual Claims, and Organizational
Forms, Harvard University Press, Cambridge, MA.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the ?rm: managerial behavior,
agency costs and ownership structure”, Journal of Financial Economics, Vol. 3 No. 4,
pp. 305-360.
Jensen, M.C. and Murphy, K.J. (1990), “Performance pay and top-management incentives”,
Journal of Political Economy, Vol. 98 No. 2, pp. 225-264.
Jensen, M.C., Murphy, K. and Wruck, E.G. (2004), “Remuneration: where we’ve been, how we
got to here, what are the problems, and how to ?x them”, Harvard NOM Working
Paper No. 04-28; ECGI – Finance Working Paper No. 44/2004, available at: http://ssrn.
com/abstract¼561305
Kent, P., Kercher, K. and Routledge, J. (2012), “Determinants of voluntary board
remuneration committee formation and composition”, paper presented at the
Accounting & Finance Association of Australia and New Zealand Conference,
Melbourne, Australia, 1-3 July 2012.
McCullagh, P. and Nelder, J.A. (1989), Generalized Linear Models, 2nd ed., Chapman & Hall,
London.
Merhebi, R., Pattenden, K., Swan, P.L. and Zhou, X. (2006), “Australian chief executive
of?cer remuneration: pay and performance”, Accounting & Finance, Vol. 46 No. 3,
pp. 481-497.
Nelder, J.A. and Wedderburn, R.W.M. (1972), “Generalized linear models”, Journal of the Royal
Statistical Society Series A (General), Vol. 135 No. 3, pp. 370-384.
Newman, H.A. and Mozes, H. (1999), “Does the composition of the compensation
committee in?uence CEO compensation practices?”, Financial Management, Vol. 28
No. 3, pp. 41-53.
Rainsbury, E.A., Bradbury, M. and Cahan, S.F. (2009), “The impact of audit committee quality on
?nancial reporting quality and audit fees”, Journal of Contemporary
Accounting & Economics, Vol. 5 No. 1, pp. 20-33.
Reynolds, J.K., Deis, D.R. Jr and Francis, J.R. (2004), “Professional service fees
and auditor objectivity”, Auditing: A Journal of Practice and Theory, Vol. 23 No. 1,
pp. 29-52.
Sarbanes-Oxley Act (2002), available at: www.pcaobus.org/ (accessed 17 July 2010).
Shaw, K.W. and Zhang, M.H. (2010), “Is CEO cash compensation punished for poor ?rm
performance?”, The Accounting Review, Vol. 85 No. 3, pp. 1065-1093.
Sheehan, K. (2009), “The regulatory framework for executive remuneration in Australia”,
Sydney Law Review, Vol. 31 No. 2, pp. 273-308.
Sheehan, K. (2012), The Regulation of Executive Compensation: Greed, Accountability and
Say on Pay, Edward Elgar Publishing Limited, Cheltenham.
Skinner, D.J. (1994), “The investment opportunity set and accounting procedure
choice: preliminary evidence”, Journal of Accounting & Economics, Vol. 16 No. 4,
pp. 407-445.
Stiglitz, J.E. (2010), Freefall: America, Free Markets, and the Sinking of the World Economy,
W.W. Norton & Co, New York, NY.
Sun, J., Cahan, S.F. and Emanuel, D. (2009), “Compensation committee governance quality, chief
executive of?cer stock option grants, and future ?rm performance”, Journal of
Banking & Finance, Vol. 33 No. 8, pp. 1507-1519.
ARJ
26,3
220
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The Corporate Law Economic Reform Program (Audit Reform and Corporate Disclosure)
Act (2004), available at: www.asic.gov.au (accessed 17 July 2010).
Tosi, H.L. Jr, Werner, S., Katz, J.P. and Gomez-Mejia, L.R. (2000), “How much does performance
matter? A meta-analysis of CEO remuneration studies”, Journal of Management, Vol. 26
No. 2, pp. 301-339.
Vafeas, N. (2003), “Further evidence on compensation committee composition as a determinant of
CEO compensation”, Financial Management, Vol. 32 No. 2, pp. 53-70.
Corresponding author
Carolyn Windsor can be contacted at: [email protected]
Remuneration
committee
independence
221
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Michael Seamer, Adrian Melia. 2015. Remunerating non-executive directors with stock options: who is
ignoring the regulator?. Accounting Research Journal 28:3, 251-267. [Abstract] [Full Text] [PDF]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_311141178.pdf
As a result of the Australian Government Productivity Commission’s recommendation to
mandate remuneration committee independence for ASX300 companies, this study aims to investigate
whether voluntary remuneration committee independence aligns chief executive officer (CEO) total
pay and bonuses with firm financial performance.
Accounting Research Journal
Remuneration committee independence and CEO remuneration for firm financial
performance
Patti Cybinski Carolyn Windsor
Article information:
To cite this document:
Patti Cybinski Carolyn Windsor , (2013),"Remuneration committee independence and CEO remuneration
for firm financial performance", Accounting Research J ournal, Vol. 26 Iss 3 pp. 197 - 221
Permanent link to this document:
http://dx.doi.org/10.1108/ARJ -08-2012-0068
Downloaded on: 24 January 2016, At: 21:18 (PT)
References: this document contains references to 63 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 1807 times since 2013*
Users who downloaded this article also downloaded:
Victoria J . Clout, Larelle Chapple, Nilan Gandhi, (2013),"The impact of auditor independence regulations
on established and emerging firms", Accounting Research J ournal, Vol. 26 Iss 2 pp. 88-108 http://
dx.doi.org/10.1108/ARJ -Dec-2011-0045
Karol Marek Klimczak, Grzegorz Szafranski, (2013),"Coincident and forecast relevance of accounting
numbers", Accounting Research J ournal, Vol. 26 Iss 3 pp. 239-255 http://dx.doi.org/10.1108/
ARJ -09-2012-0076
Prodosh Simlai, (2014),"Firm characteristics, distress risk and average stock returns", Accounting Research
J ournal, Vol. 27 Iss 2 pp. 101-123 http://dx.doi.org/10.1108/ARJ -06-2012-0046
Access to this document was granted through an Emerald subscription provided by emerald-srm:115632 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for
Authors service information about how to choose which publication to write for and submission guidelines
are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company
manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as
providing an extensive range of online products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee
on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive
preservation.
*Related content and download information correct at time of download.
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Remuneration committee
independence and CEO
remuneration for ?rm
?nancial performance
Patti Cybinski
Department of Accounting, Finance & Economics, Grif?th University Nathan,
Brisbane, Australia, and
Carolyn Windsor
Faculty of Business, Bond University, Gold Coast, Australia
Abstract
Purpose – As a result of the Australian Government Productivity Commission’s recommendation to
mandate remuneration committee independence for ASX300 companies, this study aims to investigate
whether voluntary remuneration committee independence aligns chief executive of?cer (CEO) total
pay and bonuses with ?rm ?nancial performance.
Design/methodology/approach – A series of hypotheses test the research question using multiple
regressions for a sample of 143 ASX300 companies during 2001. This time was prior to strengthen
corporate governance regulation, but after mandated executive remuneration disclosure, thus
capturing varying levels of voluntary remuneration committee independence.
Findings – This study shows ?rm size is an in?uential factor in the relationship under investigation.
ASX300 large ?rm remuneration committees link CEO total remuneration and bonuses to ?rm
?nancial performance. Smaller ASX ?rm remuneration committees do not link either type of CEO
remuneration to performance despite remuneration committee independence. Findings are mixed for
medium-sized ASX300 ?rms.
Research limitations/implications – Limitations include the necessary time restriction to 2001 for
sampling the ASX300 ?rms. The implication of this study’s ?ndings is that the proposed public policy
for mandatory remuneration committee independence is not universally effective in linking CEO
remuneration to ?rm ?nancial performance for ASX300 ?rms.
Originality/value – This study contributes to the limited research on voluntary remuneration
committee independence in relation to CEO remuneration and ?rm ?nancial performance in the
Australian context.
Keywords Chief executive of?cer, Firm ?nancial performance, Remuneration committee independence
Paper type Research paper
1. Introduction
Public anger about executive pay excesses has increased since corporate, regulatory,
and political leaders were implicated in the global ?nancial crisis that harmed
community well-being (Stiglitz, 2010). Although Australian executive pay is relatively
modest by international standards, compensation for Australian executives in the
50 to 100 largest companies has increased by as much as 300 per cent in real terms
between 1993 and 2007 (Fels, 2010). Moreover, nearly all the growth in reported
executive pay for the top 300 companies is attributed to increases in incentive pay
(Fels, 2010).
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
Accounting Research Journal
Vol. 26 No. 3, 2013
pp. 197-221
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/ARJ-08-2012-0068
Remuneration
committee
independence
197
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Australian governments however have been successfully persuaded to let the market
regulate corporate executive remuneration. Statements of good practice, such as the
ASXCorporate Governance Council’s (2003, 2007) Corporate Governance Principles and
Recommendations “Remunerate fairly and responsibly”, largely guide remuneration
practice for Australian executives. Further, government legislative intervention is most
prevalent for remuneration disclosure and shareholders’ binding[1] vote on
remuneration (Sheehan, 2009, 2012). To appease public anger, the Prime Minister
requested the Productivity Commission[2] to review the regulation of executive and
director remuneration for the following reason:
[. . .] the prime motivation for this inquiry is a widespread perception that executives have
been rewarded for failure or simply good luck. And certainly in some periods and for some
CEOs, pay outcomes appear inconsistent with a reasonably ef?cient executive labour market
(Australian Government Productivity Commission, 2009, p. xxvi).
Our study is motivated by the Commission’s recommendation to regulate the formation
and composition of independent remuneration committees for companies included in
the S&P ASX 300 index (hereafter referred to as ASX300 companies). The Commission
argues that independent remuneration committees reduce con?icts of interest. These
con?icts include executive board members who are able to make decisions about their
own pay (Australian Government Productivity Commission, 2009). This key
recommendation number (2) proposes a new ASX listing rule specifying that.
The ASX Corporate Governance Council should introduce an “if not, why not”
recommendation specifying that remuneration committees:
.
have at least three members;
.
comprise (NEDIRs), a majority of whom are independent;
.
be chaired by an independent director;
.
have a charter setting out procedures for non-committee members attending; and
.
meetings (Australian Government Productivity Commission, 2009, p. 9, xxxvii).
The research question under investigation is whether remuneration committees
composed entirely of independent directors effectively ensure that chief executive of?cer
(CEO) pay relates directly to ?rm ?nancial performance. Australian remuneration
committee composition is mainly voluntary and guided by the ASX Corporate
Governance Council (2007) best practice that formally recommends boards to establish
remuneration committees. The Council suggests that remuneration committees comprise
a majority of independent directors, chaired by an independent director with a minimum
of three members (Australian Government Productivity Commission, 2009). In 2011,
regulation of executive compensation standards has increased with the ASXintroducing
the listing requirement (ASX Listing Rule 12.8) that ASX 300 companies form a
remuneration committee comprising entirely of NEDIRs (Kent et al., 2012).
As a consequence, of the Commission’s key recommendationand the 2011 ASXlisting
rule the purpose of this study is to speci?cally examine whether independent
remuneration committees reduce con?icts of interest by effectively linking CEO pay
with ?rm?nancial performance for ASX300 companies in 2001. Since 2001 was a time of
unregulated formation and composition of remuneration committees, the regulatory
context at the time is central to this study in a similar way to Rainsbury et al. (2009)
ARJ
26,3
198
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
who examined the impact of voluntary formation of another board subcommittee, the
audit committee, in an unregulated NewZealand context. We examine the period 2001 to
capture whether ASX300 ?rm boards had voluntarily established remuneration
committees and to measure the impact of varying degrees of remuneration committee
independence. The year 2001 was just before major regulatory changes to strengthen
corporate governance, but after mandated executive remuneration disclosure in the
Australian context.
We examine this research question through a series of regression models that are
tested against their corresponding implied hypotheses about existing associations. This
study starts with the simplest association between CEOremuneration and ?rm?nancial
performance, then evolves to models that test the importance of other intervening
variables, which could impact on this association either individually or through their
interaction effects. The ?rst hypothesis examines whether a positive relationship exists
between CEO remuneration and ?rm ?nancial performance, since shareholders use
rewards to induce CEO agents to work on the owners’ behalf and reduce agency
opportunism (Fama, 1980; Fama and Jensen, 1983). A rejection of this hypothesis
indicates no evidence of CEOs being rewarded for their performance or, alternatively,
that the model is incomplete and that important intervening factors associated with the
CEO remuneration for ?rm ?nancial performance relationship have been overlooked.
Through further modelling, our study examines remuneration committee independence
and ?rm size. Interactions provide the opportunity to gain more insight about the
research question by analysing further than the initial result (Hayes and Matthes, 2009).
Since we expect that ?rm ?nancial performance does not act alone in relation to
CEO remuneration, we also include interaction effects that some of these factors may
have with the performance measure on remuneration. A two-way interaction between
?rm ?nancial performance and remuneration committee independence was ?rst
included in the model as a means of testing whether remuneration committee
independence aligns CEO remuneration with ?nancial performance. Finally, we
investigate the model with a three-way interaction to test the combined impact of ?rm
?nancial performance, remuneration committee independence, and ?rm size on CEO
remuneration since all of these factors could in?uence CEO remuneration. We then
repeat the modelling process for CEO bonuses.
This study contributes to the limited research on voluntary remuneration committee
independence, in relation to CEO remuneration and ?rm ?nancial performance, prior to
the proposed regulation of remuneration committees. Further, this study’s ?ndings
contribute towards more informed public policy associated with corporate governance
about the impact of regulation on remuneration committee formation and composition
since the Productivity Commission’s (2009) recommendation to mandate remuneration
committee independence for ASX300 companies. The Commission made its
recommendation on the basis that these ASX300 ?rms are considered large and large
?rms have the resources to establish effective remuneration committees to ensure CEOs
are paid for performance (Australian Government Productivity Commission, 2009).
We ?nd that ?rm size is signi?cantly associated with the link between CEO
remuneration, remuneration committee independence, and ?rm ?nancial performance.
The study shows, however, that ASX300 ?rms are not homogeneous in relation to
remuneration committee independence aligning CEO remuneration to ?nancial
performance. ASX300 large ?rm remuneration committees align CEO total
Remuneration
committee
independence
199
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration and bonuses to ?rm ?nancial performance. Despite their remuneration
committee independence, smaller ASX300 ?rm remuneration committees are weakly
linked to CEO remuneration including bonuses and ?nancial performance. Findings
are mixed for ASX300 medium-size ?rms with the most independent remuneration
committees aligning CEO bonuses with performance.
The remainder of the paper is organised in the following way: Section 2 discusses
studies relevant to the dependent and independent variables in the model development
process; while Section 3 explains the research design. Section 4 tests the research
question through a series of regression models and explains the analysis and Section 5
concludes the study.
2. Variables and relevant studies
2.1 Dependent variables: CEO total remuneration and bonuses
The dependent variables in this study are ASX300 Australian companies’ CEO total
remuneration and CEO bonuses in dollar terms. CEO bonuses, in particular, have
received public attention because cash incentives in the form of bonuses have grown
exponentially with more emphasis on performance-related pay for company executives
since the 1990s (Australian Government Productivity Commission, 2009). In this study,
CEOtotal remunerationcomprises the sumindollars of the ?xedcomponent (base salary,
fringe bene?ts, and superannuation) and a short-term incentive component (bonuses).
For example, the total remuneration of the Brambles Brambles, (2001) CEO comprised
salary, cashallowances, bonuses, superannuation, motor vehicle, andretirement bene?ts.
Base salary is a ?xed form of remuneration and is normally contingent on the
leadership skills and experience of CEOs. Similarly, fringe bene?ts such as motor
vehicles and superannuation payments are not contingent on performance criteria,
per se, and their relative amounts vary depending on the remuneration package
negotiated with the employer company. Jensen et al. (2004) indicate that these ?xed
forms of remuneration should not be excessive when compared to the size or
performance of the ?rm. Also, any increases in ?xed remuneration should be partly
contingent on improved ?rm ?nancial performance.
Share options are not included in this study for the following reasons, although share
options are an important element of CEO remuneration ( Jensen et al., 2004). First, CEO
total remuneration is expressed in dollar terms, which makes it dif?cult to include share
options because of incomplete and vague reporting about their dollar value. Dollar value
also varies depending on when share options have been, or will be, exercised. If share
options are not traded there is no observed market valuation that can be included in
executive pay (Hutchinson and Gul, 2004). Moreover, the features of managerial share
options are such that they have little resemblance to options on stocks or securities in
general. This also makes it dif?cult to accurately value themusing the Black andScholes
(1973) option valuation model. Second, target-based incentive plans were by far the most
widely used for Australian CEO remuneration in our 2001 sample companies
(Hay Group, 1998). Cash (rather than shares or share options) is the predominant type of
payment under these plans with the requirement of being employed at the beginning of
the period (Godfrey et al., 2003). Thus, two-thirds of our sample 2001 companies either
did not compensate or report their CEO share options.
Jensen et al. (2004) argued that there is no optimal remuneration solution as
non-economic organisational features such as culture, structure, and strategies must
ARJ
26,3
200
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
be considered. Executive remuneration is negotiated in a largely self-regulated labour
market for senior company executives, resulting in a variety of executive remuneration
packages to enhance ?exibility to suit each company’s characteristics and goals
(Sheehan, 2009). Therefore, our ?nal sample of companies varied considerably in their
remuneration structure. Not all the companies disclosed remuneration elements (such as
superannuation), while other companies did not remunerate their directors with fringe
bene?ts (such as motor vehicles). Because base salary, fringe bene?ts, and
superannuation are ?xed forms of remuneration, it was possible to combine the three
forms of compensation into a total remuneration package. For this study, we examined
CEO total remuneration and CEO bonuses as dependent variables.
2.2 Independent variables
In the following section, we discuss the relevant research associated with the
independent variables that are linked to CEO remuneration and bonuses including ?rm
?nancial performance, remuneration committee independence, and ?rm size.
2.2.1 Firm ?nancial performance measure. The UK Cadbury Committee report
(Cadbury, 1992) was one of the earliest investigations into the integrity of corporate
governance. This committee recommended a voluntary code of best practice to include
a remuneration committee, as described by Ezzamel and Watson (1998, p. 222):
The remuneration of executive directors (those who are executives of the company on whose
board they sit) should be subject to the recommendations of a remuneration committee
(made up wholly or mainly of non-executive directors), which should ensure that large pay
awards were justi?ed by increases in ?rm performance and shareholders’ wealth.
Corporate boards argue that senior executives, particularly the CEO, should be
rewarded for managing a pro?table company that provides appropriate returns to the
owners, and, at the same time, for reducing the agency problem or management
opportunism (Fama and Jensen, 1983; Fama, 1980). From a managerialism perspective,
Berle and Means (1932, p. 25) identi?ed ?aws in the separation of ownership and
control, stating that “the separation of ownership from control produces a condition
where the interests of owner and of ultimate manager may, and often do, diverge, and
where many of the checks which formerly operated to limit the use of power disappear”
(cited Tosi et al., 2000, p. 302). Managerialist logic argues that without external
constraints, executives are more focused on increasing the ?rm size to increase their
pay rather than increasing pro?ts and ?rm performance (Tosi et al., 2000).
The separation of ownership from control inspired agency theory ( Jensen and
Meckling, 1976), which is based on the premise that principals (the shareholders)
delegate duties to an agent (the CEO), who is expected to act in the interest of the
principal. Agency theory is widely applied in corporate governance research because it
simpli?es assumptions about the two participating parties, shareholders and managers
whose interests are assumed to be clear and consistent (Daily et al., 2003). Agency theory
makes assumptions about managers’ behaviour: to illustrate, agents are assumed to
be risk averse and self-centred and therefore, agents’ interests may differ from those of
the principal. Opportunistic behaviour by the agent is possible as the agent may have
different objectives from the principal and, thus, pursue a self-serving agenda. The
board, through the remuneration committee, use a variety of incentives such as bonuses
and options in an attempt to align management pay with ?rm ?nancial performance in
the interest of shareholders ( Jensen, 2000; Jensen et al., 2004).
Remuneration
committee
independence
201
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Evidence is mixed regarding the link between CEO remuneration and ?rm
?nancial performance. While Merhebi et al. (2006) found a positive relationship
between CEO remuneration and ?rm ?nancial performance in their Australian study,
earlier Australian research for the years 1987-1992, by Izan et al. (1998), reported
no link between CEO remuneration and ?rm ?nancial performance. A later study
that examined Australian board structural independence (1999-2006) reported that
independent boards were no better than executive dominated boards in enforcing CEO
remuneration for ?nancial performance (Capezio et al., 2011).
Shaw and Zhang (2010) argued, however, that remuneration committees are reluctant
topunishCEOremunerationfor poor earnings, whenthose poor earnings result, inpart, on
investment decisions that are expected to generate positive future returns. Examples of
positive investment decisions are research and development expenditures or
restructurings. Using a yearly change in return on assets (ROA) as one proxy for
performance, Shaw and Zhang (2010) found in their US study that punishment in CEO
cash compensation for poor ?rm ?nancial performance does not systematically occur.
Their research also examined the CEO remuneration-performance relation for ?rms with
strong corporate governance. Their results provided further doubt on the role of ex post
settlingupinCEOcashcompensation. Inaddition, theyinvestigatedthe role of accounting
earnings in CEO compensation plans. Shaw and Zhang (2010) found that CEO
remuneration-performance sensitivity weakens for poor earnings performance, which is
consistent with studies that ?nd CEO cash compensation is shielded from transitory
accountinglosses (Gaver andGaver, 1998) andrestructuringcharges (Dechowet al., 1994).
ROA is, therefore, the primary measure of company ?nancial performance in this
study because companies typically base the incentive components of CEOremuneration
on ?nancial accounting based measures (Merhebi et al., 2006; Shaw and Zhang, 2010).
Further, the Australian Productivity Commission (2009, p. 70) stated that:
Researchers investigating the relationship between pay and corporate performance are
limited to using publicly-available indicators. Some have used regression analysis to identify
statistical relationships between indicators of corporate performance and executive
remuneration. This approach is reasonable provided the data are adequate and the
indicators correspond with those used by companies to determine executive remuneration.
In this study, we lag ROA by one year as we expect CEO compensation to be based on
?rm ?nancial performance for the prior year.
2.2.2 Remuneration committee independence. Some executives sit on boards, thus
creating con?icts of interest when setting executive pay. To reduce con?icts of interest
the Cadbury Committee (1992) recommended that the remuneration committee comprise
a majority of NEDIRs to enhance independent decision-making about executive pay.
Remuneration committees generally consist of independent NEDIRs to form an arm’s
length group to enhance the integrity of the decision-making process. The Corporations
Act 2001 (Cth) does not de?ne independence of directors, and Australian boards
generally adopt their own de?nition. The ASX Corporate Governance Council (2007)
de?nes an independent NEDIRs as one “who is free of any business or other
relationship that could materially interfere with the independent exercise of their
judgment, or be perceived to do so” (as cited in Australian Government Productivity
Commission, 2009, p. 174).
In 1993, independent remuneration committee composition became core US
regulation introduced by the Securities and Exchange Commission (SEC) and the
ARJ
26,3
202
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Internal Revenue Service (Vafeas, 2003). These regulations were motivated by the
concern that when executive directors participated in remuneration committees’
executive pay contracts, the remuneration committee was compromised resulting in pay
contracts that favoured management (Vafeas, 2003).
Vafeas (2003) investigated the impact of the US regulations on the requirement of
remuneration committees to comprise a majority of NEDIRs. After analysing 271 ?rms
for the 1991-1997 period, he reported a declining trend in executive director
membership of remuneration committees However, based on the evidence, he could not
conclude de?nitively that shareholders bene?ted from the US regulations to enhance
remuneration committee independence. Vafeas (2003) suggested that ?rms with
remuneration committees comprising outsiders were able to fend off uninvited public
scrutiny of pay practices and complying with the SEC’s 1992 disclosure rules about the
governance of executive compensation.
Conyon and Peck (1998), in their UK study of remuneration committees and executive
pay for 94 companies in the period 1991-1994, reported the positive relation between the
proportion of NEDIRs on a remuneration committee and senior management pay and
sensitivity of pay to performance. In contrast, Daily et al. (1998) found no link between
excessive CEO remuneration and remuneration committee dominance by executive
directors in their study of 194 US?rms for the period of 1991. These ?ndings are supported
by Newman and Mozes (1999), who analysed 1992 US company data for 161 ?rms and
found no relationship between CEO remuneration and executive director participation in
the remuneration committee. However, they noted that under certain conditions executive
pay for performance is skewed in management’s favour.
Sun et al. (2009) argued that because the SEC rules mandated wholly independent
compensation committees for US listed companies in 2003, a new and
more comprehensive measure of compensation committee quality was required. To
measure compensation committee quality, in relation to stock option grants and future
?rm performance, their study introduced and factor-analysed the following six
characteristics: CEO appointed directors, long-serving outside directors, CEOs from
other ?rms, high director stock ownership, busy outside directors (with three or more
additional board seats), and larger boards.
This study concentrates on remuneration committee independence, as the Australian
Government Productivity Commission (2009, p. xxxvii) proposed public policy to
mandate remuneration committee independence stated in their Recommendation 2 for
ASX300 ?rms. The aim of this study, therefore, is to investigate whether remuneration
committees comprising NEDIRs link CEO remuneration and bonuses with performance
of Australian companies for 2001. In our study, the proportion of NEDIRs comprising the
remuneration committee measures independence, de?ned by the ASX Corporate
Governance Council (2003, 2007), and is categorised into three levels for the regression
analyses that included factor interactions (Table I and II).
2001 company reports ASX300
Total ?rms in the dataset 300
Less trusts and unavailable ?rms 2001 reports due to
insolvencies, mergers, or no longer trading on ASX
(124)
Less ?rms with incomplete data (33)
Total sample 143
Table I.
Remuneration
committee
independence
203
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2.3 Firm size and executive compensation. Firm size does in?uence executive pay, as
larger companies seem to be prepared to pay for increased job importance and
complexity to attract more talented people, thus explaining 25-50 per cent observed
increases in executive pay in larger ?rms (Australian Government Productivity
Commission, 2009). Tosi et al. (2000) examined the links between ?rm size, ?rm
performance, and CEO remuneration in their meta-analysis, which formed part of
a review of the empirical literature on the determinants of CEO remuneration. Their
hypotheses speci?cally focused on:
.
?rm performance and CEO remuneration; and
.
?rm size and CEO remuneration.
They found that ?rm size accounted for more than 40 per cent of the variance in total
CEO remuneration, while ?rm performance accounted for less than 5 per cent of the
variance. Merhebi et al. (2006) tested the link between ?rmsize and CEOremuneration of
Australian companies (1990-1999) and found a signi?cant and positive result. They
argued that some evidence exists regarding size as a proxy for performance and that
“larger ?rms clearly have the capacity for higher remuneration packages regardless of
performance” (Merhebi et al., 2006, p. 495). Baker et al. (1988) questioned the notion of
size as a proxy for performance when they found evidence that CEOs are able to increase
their pay by increasing the ?rm size, even when the size increase reduces the ?rm’s
market value. They further suggested this motivation “could explain some of the vast
amount of inef?cient expenditures of corporate resources on diversi?cation programs
that have created large conglomerate organizations” (Baker et al., 1988, p. 609).
Additionally, Tosi et al. (2000, p. 329) argued that their ?ndings were consistent with:
[. . .] those theoretical explanations that emphasize organizational size as an important
determinant of total CEO remuneration; that is, indicators of ?rm size, taken together, explain
almost nine times the amount of variance in total CEO remuneration than the most highly
correlated performance measure. A lesser effect is demonstrated in the ?ndings regarding
pay sensitivity as well as in the difference in the pay/performance or pay/?rm growth
sensitivities. Changes in ?rm performance account for only 4 per cent of the variance in CEO
remuneration, while changes in ?rm size account for 5 per cent of the variance in CEO
remuneration. These results are consistent with Jensen and Murphy’s (1990) conclusion that
“incentive alignment” as an explanatory agency construct for CEO remuneration is weakly
supported at best.
Anumber of measures of ?rmsize appear in the literature, including sales revenue, log of
net sales, net income, total assets (Hagerman and Zmijewski, 1979; Skinner, 1994), and
log of total assets (Reynolds et al., 2004). Hagerman and Zmijewski (1979) argue that no
measure of size is necessarily better than another. Therefore, we use total assets as a
proxy for ?rm size and partition our sample into thirds for large, medium, and smaller
?rms to normalise the data (see Table III for mean total assets for each category).
3. Research design
Sample
To capture the varying levels of voluntary remuneration committee independence, we
expressly analyse data from ASX300 companies’ 2001 annual reports. The timeframe,
2001, is of interest because it is just after mandated executive remuneration disclosure,
but before the introduction of regulation to strengthen corporate governance including
ARJ
26,3
204
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the ASX recommendation for remuneration committee independence. In fact, little
variation now exists in remuneration committee composition of ASX300 companies as
most now have independent remuneration committees. Further, Clarkson et al. (2011)
concluded that extensive regulatory change between 2001 and 2009 had enhanced
oversight of the executive compensation process through strengthened corporate
governance regulations and increased transparency.
Overview of increased corporate governance regulation post-2001
We, therefore, examine ASX300 companies in a corporate governance context before the
enactment of legislation that increased regulatory scrutiny of directors’ responsibilities
in response to corporate collapses, such as Enron in December 2001. The US Congress
enhanced corporate regulation by enacting the Sarbanes-Oxley Act (SOX) (2002). These
new corporate governance rules and requirements affected US-based trading ?rms and
Australian companies associated with the US capital markets. Additionally, Australian
based transnational audit ?rms began to use elements of SOX in their audits of
Australian companies. The Australian Government’s Company Law Economic Reform
Program (CLERP), concerning corporate governance, soon followed and was enacted
as CLERP 9 in 2004. Moreover, the ASX Corporate Governance Council’s (2003, 2007)
Corporate Governance Principles and Recommendations were introduced and
encouraged ?rms to implement independent remuneration committees that included a
majority of NEDIRs.
Mandated executive remuneration disclosure pre-2001
Australian regulation mandated disclosure of directors’ and executives’ remuneration
with the introduction of the Company Law Review Act 1998, revised May 1999
(Australian Securities and Investments Commission, 1998). Section 300A required
executive remuneration to be disclosed by listed Australian companies for ?nancial
years after 1 July 1998. The Act requires ?nancial statement disclosure of emoluments
of the ?ve most highly remunerated of?cers. Originally, executive remuneration was
disclosed in bands and not the actual amounts paid to executives. This requirement
was removed and the accounting standard required companies to disclose aggregate
remuneration of all executive of?cers whose remuneration for the ?nancial year is
$100,000 (Australian Accounting Standards Board (AASB) (1999) 1034).
The amended AASB 1034 became operative for companies at ?nancial year end
30 June 2001, so our sample of companies covers this speci?c period for the end of
the ?nancial year 2001. AASB (2005) 101 superseded this standard in July 2004
Level n
REMCindep
0 Lowest proportion: ?rms with no remuneration
committee or a minority of NEDIRs – least
independent
27
1 Firms with a majority of NEDIRs on remuneration
committees
42
2 100 percent NEDIRs on remuneration committees –
most independent
74
Total 143
Table II.
Remuneration committee
independence as
proportion of NEDIR at
three levels
Remuneration
committee
independence
205
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(operative 1 January 2005) to comply with the International Accounting Standards and
executive remuneration was speci?cally guided by AASB (2004) 1046 Director and
Executive Disclosures by Disclosing Entities.
Events in?uencing ?nal sample
Other contextual events in?uenced the collection of our sample of 2001 company
?nancial reports. These events include the internet bubble crash in March 2001 that
resulted in several corporate collapses and general investor turmoil in the following
years’ ?nancial reports (Boyer, 2005). For example, we looked for several IT companies
that were in our 2001 sample, but found they were non-existent in 2002, either insolvent
or merged. The 11 September 2001 attack on New York’s Twin Towers also resulted in
further investor turmoil that exacerbated the impact of the internet bubble crash.
Additionally, we found ASX300 company composition after 2001 was affected by high
changeability (survivorship) due to mergers, insolvencies, and, to some extent, the
re-privatising of public companies, making it dif?cult to sustain a longitudinal study
with a reasonable sample of the original 2001 ASX300 companies.
Our ?nal sample of companies in this study consists of 143 large publicly listed
Australian companies. The companies were chosen from the S&P/ASX 300[3] database
for 2001. Given that compulsory disclosure of corporate governance practices and
executive remuneration has only existed since late 1998 in Australia, extending the scope
of this study prior to this period would be pointless, as appropriate data to carry out the
study would not be publicly available. Further, we concluded that the effectiveness of
various committees, suchas remunerationcommittees, couldnot be evaluatedbefore 2000
because these committees were in the process of being established in many companies.
We individually downloaded S&P/ASX300 company reports for 2001 from
FinAnalysis, supplemented with data from Connect4 databases. From the 300 ?rms,
the sample consisted of 176 ?rms after excluding 124 ?rms that comprised trusts and
?rms whose reports were unavailable owing to insolvencies, mergers or no longer
trading. Out of the initial 176 companies, a ?nal sample of 143 companies had the
appropriate CEOremuneration and remuneration committee information. Surprisingly,
our sample included 45 companies that chose not to compensate their CEOs with
incentives. This is contrary to the recommendations of various committees, such as the
Cadbury Committee, and previous research (Baber et al., 1996; Beatty and Zajac, 1994;
Brunello et al., 2001; Carpenter and Sanders, 2002) which suggests that an effective
mechanism for aligning management’s interest with that of shareholders is to
compensate managers with short- and long-term incentives that are linked to ?rm
?nancial performance. Table I describes the sample used in this study.
Each report in our ?nal sample of 143 was manually examined to collect information
pertinent to this study. This hand-collected data includedthe dollar amounts for the CEO
remuneration components required for the dependent variables. The proportion of
NEDIRs on the remuneration committee was also hand collected for the categorical
independent variable remuneration committee independence. Essentially, the sample
companies had to disclose base salary, superannuation, fringe bene?ts, bonuses, share
options, and corporate governance information about the composition of remuneration
committees. FinAnalysis provided data for accounting information, ROA, and dollar
value of total assets to measure the independent variables, ?rm ?nancial performance,
and ?rm size.
ARJ
26,3
206
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
3.1 Measurement of variables
CEOtotal is the dependent variable measured as the natural logarithm of CEO
remuneration comprising the dollar sum of base pay, superannuation, fringe bene?ts,
and bonuses reported in 2001 ASX300 company reports.
CEObonus is the other dependent variable measured as the natural logarithm of the
short-term incentive for performance (in dollars) as part of CEO total remuneration
from the 2001 ASX300 company reports.
ROAlag1 is a continuous independent variable that is a measure of company
performance resulting from management’s productive use of company assets. ROA is
calculated as earnings before interest (total assets less outside equity interests) and is a
key measure of a company’s pro?tability, equal to a ?scal year’s earnings divided by its
total assets. ROAessentially shows howmuch pro?t a company is making on the assets
used in its business. ROA is lagged, as CEO compensation in our 2001 sample would
have been based on ?rm ?nancial performance for the prior year (2000).
REMCindep is measured as a proportion (percentage) of NEDIRs in the committee.
It was calculated by summing the number of NEDIRs in the committee and dividing
that ?gure by the total number of directors in that committee. Table II reports the
categorisation of REMCindep into three levels as follows: companies with no
remuneration committees or remuneration committees with a minority proportion of
(NEDIRs) are designated as 0; companies with a majority proportion less than 100 per
cent of NEDIRs on the remuneration committee are designated as 1; and, those
companies having all NEDIRs on the remuneration committee are designated as 2.
Size is an independent variable measured as total assets ($ value) categorised into
three levels of approximately equal number of ?rms: smaller ?rms ¼ 0, medium
?rms ¼ 1, and larger ?rms ¼ 2 (with the mean total assets for each category shown in
Table III). Transformation of ?rm Size into log total assets is unnecessary as this is
a categorical variable and, therefore, treated as a categorical variable in the
regression models.
Industry is a categorical variable to control for industry differences. Industry groups
comprise the ten sectors of the Global Industry Classi?cation Standard (GICS) used by
the ASX for industry classi?cation (Australian Securities Exchange, 2010a, b). For
analysis purposes, this study classi?es ten industry sectors into four industry groups
based on similarities in the nature of the industry. As outlined in Table IV, Panels Aand
B, industry is categorised: energy and mining ¼ 1, industrials ¼ 2, consumer ¼ 3, and
services ¼ 4.
4. Model development and results
Through a series of evolving models, we test the research question: whether
remuneration committee independence (REMCindep) effectively links CEO total
Size at three levels Level Mean of assets $M SD n
Smaller ?rms 0 148 779 46
Medium-sized ?rms 1 787 37 49
Larger ?rms 2 18,980 47,950 48
Total mean 143
Table III.
Firm size (total asset –
$ millions) at three levels
Remuneration
committee
independence
207
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration (CEOtotal ) and bonuses (CEObonus) with prior year’s ?rm ?nancial
performance (ROAlag1) for the period of 2001.
A higher ROA indicates management’s ability to use company assets ef?ciently in
serving shareholders’ economic interests. Therefore, we expect a positive association.
The link between CEO total remuneration and ?rm ?nancial performance is ?rst tested
without taking into consideration the effect of remuneration committee independence.
We use the general linear model (GLM) procedure in SPSS. This is a ?exible statistical
model that incorporates normally distributed dependent variables and categorical or
continuous independent variables (McCullagh and Nelder, 1989; Nelder and
Wedderburn, 1972). GLM allows factors (and interactions of factors) as predictors,
rather than just continuous variable predictors, in the models we are estimating; and is
appropriate because this research relied on considering the interactions between factor
variables.
Model 1: regression – CEO total remuneration and ?rm ?nancial performance
CEOtotal ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
Model 1 showed a non-signi?cant relationship for the link between CEO
remuneration and ?rm ?nancial performance (Table V, p ¼ 0.977). We argue,
however, that Model 1 is incomplete and other important factors are involved in
explaining CEO remuneration outcomes. Other factors deemed important in the
literature (i.e. remuneration committee independence and ?rm size) are tested by
considering Model 2.
Model 2 examines and tests the impact of an independent remuneration committee
as a corporate governance intervention. The independence of the remuneration
committee is represented by the proportion of NEDIRs on that committee, categorised
into three levels as follows: 74 ?rms had remuneration committees comprising
100 per cent of NEDIRs (i.e. the most independent remuneration committees); 42 ?rms
had remuneration committees with majority NEDIR composition; and 27 ?rms had no
Panel A: industry groups based on GICS code and industry sector
Industry groups GICS code Sector
1. Energy and mining 10 Energy
15 Materials
2. Industrials 20 Industrials
3. Consumer 25 Consumer discretionary
30 Consumer staples
4. Services 35 Health care
40 Financials
50 Telecommunication services
55 Utilities
Panel B: the no. of ?rms for each industry group
Industry sector
1. Energy and mining 33 (23.132%)
2. Industrials 21 (14.673%)
3. Consumer 44 (30.674%)
4. Services 45 (31.521%)
Table IV.
ARJ
26,3
208
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
remuneration committees or remuneration committees with a minority of NEDIRs
(i.e. the least independent remuneration committees, see Table II).
Model 2: regression, a two-way interaction model for CEO total remuneration on ?rm
?nancial performance and remuneration committee independence
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
where: b1 is now a vector of coef?cients for the various levels of the interaction term in
the model.
In Model 2, we test the effect of a two-way interaction between ?rm ?nancial
performance and remuneration committee independence on CEO remuneration. This is
to investigate whether remuneration committees control CEO pay in line with ?rm
?nancial performance in at least some partitions of the group of ?rms characterized by
the different levels of remuneration committee independence. For theoretical reasons,
the main effects ROAlag1 and REMCindep are not included because the interaction
effect between themis signi?cant in Model 2 ( p ¼ 0.01). To set up a model that includes
these two main effects, and interpret their coef?cients and hypothesis tests as main
effects, is not theoretically justi?ed or correct when their interaction effect is signi?cant,
except under limited conditions (Hayes, 2005, pp. 452-456; Irwin and McClelland, 2001;
Jaccard and Turrisi, 2003, p. 24).
The two-way interaction model for CEOtotal remuneration is signi?cant (F
3,136
¼ 3.7;
p ¼ 0.01, Table VI) indicating that independent remuneration committees are
signi?cantly associated with the relationship between CEO total remuneration and
prior year’s ?rm ?nancial performance. We examine the three coef?cient parameter
estimates for the interaction term in Table VII to ascertain in which subgroup/s of
remuneration committee independence the association between CEO remuneration and
?rm?nancial performance is signi?cant, and whether it is a positive or negative one. We
?nd that for the most independent remuneration committees (with 100 per cent NEDIR
composition) the association between performance and CEO remuneration is signi?cant
and positive ( p ¼ 0.06, Table VII). However, for the least independent remuneration
committees (with either no remuneration committee or a minority of NEDIRs) the
association between ?rm?nancial performance and CEOremuneration is signi?cant and
negative ( p , 0.05, Table VII).
Variable F p-value
Intercept 26,256.166 0.000
Industry 0.147 0.932
ROAlag1 0.004 0.952
Likelihood ratio x
2
¼ 0.462, df ¼ 4 ( p ¼ 0.977)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; industry – industry
groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table V.
Model 1 CEO total
remuneration and ?rm
?nancial performance
Remuneration
committee
independence
209
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This explains the lack of signi?cance for Model 1, since the CEO remuneration for ?rm
?nancial performance relationship is neither consistently signi?cant nor consistently
positive for all three levels of remuneration committee independence. The model ?t is
signi?cant ( p ¼ 0.07 , 0.1), using the maximum likelihood ratio x
2
test. However,
the model ?t can be improved by including ?rmsize as a three-way interaction termwith
the other two independent variables.
Variable b SE t p-value
Intercept 13.67 0.158 86.467 0.000
Energy and mining 20.056 0.210 20.265 0.792
Industrials 20.163 0.241 20.675 0.501
Consumer 20.125 0.193 20.648 0.518
Services 0
a
ROAlag1
*
[REMCindep ¼ 0] 20.024 0.012 21.996 0.048
* *
ROAlag1
*
[REMCindep ¼ 1] 0.026 0.018 1.449 0.150
ROAlag1
*
[REMCindep ¼ 2] 0.025 0.013 1.883 0.062
*
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithmof CEOtotal remuneration ($) comprisingbase pay þ superannuation/fringe
bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; REMCindep is the proportion of
NEDIRs onthe remuneration committee (RC), categorised into 0 – ?rms with noRCs or a minorityNEDIRs,
the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but less than 100 per cent; 2 – 100 per cent
NEDIR composition, the most independent RCs; industry – industry groups based on GIC sectors, 1 –
energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VII.
Model 2 CEO total
remuneration parameter
estimates for two-way
Interaction between ?rm
?nancial performance
and remuneration
committee independence
Variable F p-value
Intercept 2,624.072 0.000
Industry 0.216 0.885
ROAlag1
*
REMCindep 3.717 0.013
* * *
Likelihood ratio x
2
¼ 11.7, df ¼ 6 ( p ¼ 0.068)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithmof CEOtotal remuneration ($) comprisingbase pay þ superannuation/fringe
bene?ts þ bonuses; ROAlag1 – return on assets for the prior year 2000; REMCindep is the proportion of
(NEDIRs) ontheremunerationcommittee (RC), categorisedinto0 – ?rms withnoRCs or aminorityNEDIRs,
the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but less than 100 per cent; 2 – 100 per cent
NEDIR composition, the most independent RCs; industry – industry groups based on GIC sectors, 1 –
energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VI.
Model 2 CEO total
remuneration and
two-way interaction
between ?rm
performance and
remuneration committee
independence
ARJ
26,3
210
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Model 3: regression – a three-way interaction model for CEO total remuneration on
?rm ?nancial performance, remuneration committee independence and ?rm size
Prior research by Tosi et al. (2000) and Merhebi et al. (2006) examined ?rm size and
?nancial performance separatelyinrelationto CEOremuneration. These studies suggest
that large ?rms generously pay their CEOs and executives and size was found to be a
more important criterion than ?nancial performance alone. Recall that our study
contributes to corporate governance research and associated public policy by examining
whether varying levels of remuneration committee independence align CEO total
remuneration with ?rm ?nancial performance for ASX 300 companies in an essentially
voluntary context. Although we examine ASX300 ?rms, these ?rms vary in size and are
not homogeneous. Firmsize (Size) is categorisedinto the followingthree levels (described
in Table III): smaller ?rms (n ¼ 46, mean 148, SD77 in $m), medium?rms (n ¼ 49, mean
787, SD 373 in $m), and large ?rms (n ¼ 48, mean 18,980, SD 47,950 in $m). This
categorisation of size allows analysis of the model with a three-way interaction below:
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
where: b1 is a vector of coef?cients for the various levels of the interaction term in the
three-way interaction model (Friedrich, 1982; Jaccard and Turrisi, 2003). For Model 3,
Table VIII reports a signi?cant high-order relationship (F
12,130
¼ 8.94, p ¼ 0.001)
between the dependent variable, CEO total remuneration (CEOtotal ) and the three-way
interaction with ?rm size (Size), prior year’s ?rm ?nancial performance (ROAlag1), and
remuneration committee independence (REMCindep). Model 3 is a much better ?t than
Model 2, as indicated by the maximum likelihood x
2
test ( p , 0.001). Table IX reports
further analysis of this three-way interaction revealing that ?rmsize matters, with very
signi?cant positive associations for large ?rms ( p , 0.01) between CEOtotal and ?rm
?nancial performance (ROAlag1) for all three levels of remuneration committee
independence (REMCindep).
The results for the three-way interaction are particularly strong ( p , 0.001),
especially for large ?rms with remuneration committees comprising 100 per cent
Variable F p-value
Intercept 12,175.206 0.000
Industry 1.469 0.226
Size
*
ROAlag1
*
REMCindep 8.940 0.000
* * * *
Likelihood ratio x
2
¼ 69.353, df ¼ 12 ( p ¼ 0.000)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are
two-tailed):
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; Size – total assets ($ value) categorised into thirds: 0 – smaller ?rms, 1 –
medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000; REMCindep is the
proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms with no RCs or a
minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs but less than
100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; industry – industry
groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table VIII.
Model 3 CEO total
remuneration and
three-way interaction
between ?rm size, ?rm
?nancial performance
and remuneration
committee independence
Remuneration
committee
independence
211
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
NEDIRs and for ?rms with a majority of NEDIRs ( p ¼ ,0.001). Results are weaker
( p , 0.005) for ?rms with no remuneration committees or a minority NEDIRs. For
medium and smaller size ?rms, the only signi?cant positive association between CEO
total pay and prior year’s ?rm ?nancial performance occurred for medium-sized ?rms
with remuneration committees comprising a majority of NEDIRs ( p , 0.05).
Interestingly, a signi?cant negative relationship between CEOtotal and prior year’s
?rm ?nancial performance was found for smaller ?rms with no remuneration
committees or remuneration committees with a minority of NEDIRs (the lowest level of
remuneration committee independence). This last result suggests that smaller ?rms
with the least independent remuneration committees recompense their CEOs even when
these ?rms are not performing ?nancially. The results also show the ineffectiveness of
remuneration committees to align CEO total pay with ?rm ?nancial performance for
smaller ?rms regardless of the composition of the remuneration committee.
4.1 CEO bonuses
CEO bonuses represent short-term incentives that should be based directly on ?rm
?nancial performance. Therefore, similar to previous arguments, an independent
remuneration committee should control CEOincentive pay or bonuses for performance.
The testing process of Models 1-3 is a similar process as for Models 4-6, but with
CEO bonuses as the dependent variable.
Variable b SE t p-value
Intercept 13.554 0.141 96.452 0.000
Energy and mining 20.312 0.186 21.680 0.095
Industrials 20.171 0.204 20.838 0.404
Consumer 20.324 0.170 21.908 0.059
Services 0
a
[Size ¼ 0 smaller]
*
[REMCindep ¼ 0]
*
ROAlag1 20.023 0.011 22.184 0.031
* *
[Size ¼ 0 smaller]
*
[REMCindep ¼ 1]
*
ROAlag1 20.016 0.021 20.751 0.454
[Size ¼ 0 smaller]
*
[REMCindep ¼ 2]
*
ROAlag1 0.003 0.013 0.257 0.798
[Size ¼ 1 med]
*
[REMCindep ¼ 0]
*
ROAlag1 20.014 0.030 20.467 0.641
[Size ¼ 1 med]
*
[REMCindep ¼ 1]
*
ROAlag1 0.059 0.025 2.335 0.021
* *
[Size ¼ 1 med]
*
[REMCindep ¼ 2]
*
ROAlag1 0.032 0.025 1.311 0.192
[Size ¼ 2 large]
*
[REMCindep ¼ 0]
*
ROAlag1 0.152 0.051 2.970 0.004
* * *
[Size ¼ 2 large]
*
[REMCindep ¼ 1]
*
ROAlag1 0.126 0.025 4.991 0.000
* * * *
[Size ¼ 2 large]
*
[REMCindep ¼ 2]
*
ROAlag1 0.139 0.023 6.153 0.000
* * * *
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEOtotal ¼ b0 þb1ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEOtotal – natural logarithm of CEO total remuneration ($) comprising base pay þ superannuation/
fringe bene?ts þ bonuses; Size – total assets ($ value) categorised into thirds: 0 – smaller ?rms, 1 –
medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000; REMCindep is the
proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms with no RCs or a
minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs but less than 100
per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; industry – industry groups
based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 – services
Table IX.
Model 3 CEO total
remuneration parameter
estimates for three-way
interaction between ?rm
size, remuneration
committee independence,
and ?rm ?nancial
performance
ARJ
26,3
212
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Model 4: regression – CEO bonuses and ?rm ?nancial performance
CEObonus ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
Model 4, with CEO bonuses (CEObonus) as the dependent variable, is not signi?cant
( p ¼ 0.22) indicating that our sample of ?rms showed no evidence of aligning CEO
bonus pay with prior year’s performance (Table X). As before, we argue that this model
is incomplete and other intervening variables are involved. Thus, we include
remuneration committee independence in Model 5 to improve the model ?t, as in
Models 1-3 that tested CEO total remuneration as the dependent variable.
Model 5: regression – a two-way interaction model for CEO bonuses on ?rm ?nancial
performance and remuneration committee independence
CEObonus ¼ b0 þb1ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
In Model 5 we test CEO bonuses and the two-way interaction between ?rm ?nancial
performance and remuneration committee independence. The result was not
signi?cant ( p ¼ 0.238), as shown in Table XI. The two-way interaction shows only
marginal signi?cance ( p ¼ 0.1). We conclude that our sample of ?rms, when
partitioned by the independence level of their remuneration committee, show no
evidence of aligning CEO bonuses with prior year’s ?rm ?nancial performance within
any subgroup of remuneration committee categorised at three levels of independence.
Model 6: regression model – a three-way interaction model for CEO bonus on ?rm
?nancial performance, remuneration committee independence and ?rm size
Next, in Model 6, we investigate the association between the dependent variable CEO
bonus, and the three-way interaction between ?rm size, prior year’s ?rm ?nancial
performance, and remuneration committee independence. Table XII Model 6 reports a
signi?cant three-way interaction (F
9,129
¼ 2.24, p , 0.05), but not as strong as for
Model 3 that shows CEO total remuneration ( p , 0.001). Just as for CEO total
remuneration, Table XIII reports further analysis of Model 6 interaction and shows a
strong positive relationship between CEO bonuses and prior year’s ?nancial
performance (ROAlag1) for large ?rms at all three levels of remuneration committee
independence (REMCindep). The relationship is strongest ( p , 0.002) for large ?rms
Variable F p-value
Intercept 102.202 0.000
Industry 1.653 0.180
ROAlag1 0.589 0.444
Likelihood ratio x
2
¼ 5.73, df ¼ 4 ( p ¼ 0.22)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ROAlag1 þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; ROAlag1 – return on assets for the prior year 2000;
industry – industrygroups basedonGICsectors, 1 – energyandmining, 2 – industrials, 3 – consumer and
4 – services
Table X.
Model 4 CEO bonuses
with ?rm ?nancial
performance
Remuneration
committee
independence
213
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
with remuneration committees comprising 100 per cent NEDIRs. This result shows
that the largest ?rms with the most independent remuneration committees consistently
paid CEO bonuses in line with ?rm ?nancial performance, thus demonstrating that
CEO bonuses are somewhat controlled by independent remuneration committees for
this subgroup of ?rms.
We ?ndaweaker result for large ?rms withremunerationcommittees havinga majority
of NEDIRs or mid-level independence ( p , 0.05); while, those ?rms with no remuneration
committees or a minority of NEDIRs were weaker still ( p , 0.01). Table XIII also reports a
positive signi?cant result for medium-sized ?rms with 100 per cent NEDIRs ( p ¼ ,0.05).
These results suggest that large and medium ?rm groups with 100 per cent NEDIRs on
their remuneration committees are the most effective in controlling CEO bonus incentives
in line with prior year’s ?rm ?nancial performance. The results also show the
Variable F p-value
Intercept 40.354 0.000
Industry 2.560 0.058
Size
*
ROAlag1
*
REMCindep 2.244 0.023
* *
Likelihood ratio x
2
¼ 25.7, df ¼ 12 ( p ¼ 0.012)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; Size – total assets ($ value) categorised into thirds: 0 –
smaller ?rms, 1 – medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year 2000;
REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms
with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rm RCs
but less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; Industry –
industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 –
services
Table XII.
Model 6 CEO bonuses
and three-way interaction
between ?rm size, ?rm
?nancial performance,
and remuneration
committee independence
Variable F p-value
Intercept 106.083 0.000
Industry 1.703 0.169
ROAlag1
*
REMCindep 2.727 0.101
Likelihood ratio x
2
¼ 8.0, df ¼ 6 ( p ¼ 0.238)
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed):
CEObonus ¼ b0 þ b1 ðROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; ROAlag1 – return on assets for the prior year 2000;
REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0 – ?rms
with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on ?rmRCs but
less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs; Industry –
industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 – consumer and 4 –
services
Table XI.
Model 5 CEO bonuses
and two-way interaction
between ?rm ?nancial
performance and
remuneration committee
independence
ARJ
26,3
214
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ineffectiveness of remuneration committees to align CEO bonuses with ?rm ?nancial
performance for the smaller ?rms regardless of the composition of the remuneration
committee.
5. Discussion and conclusion
The Productivity Commission’s key draft recommendation (Australian Government
Productivity Commission, 2009, p. xxxiii) proposed a new ASX listing rule specifying
that “all ASX300 companies have a remuneration committee of at least three members,
all of whom are non-executive directors, with the chair and a majority of members
being independent” to avoid con?icts of interest when paying executives. Our study
examines whether remuneration committee independence aligns CEO compensation
with performance for ASX300 ?rms during 2001, a period of voluntary corporate
governance.
We show that ?rm size is an important factor in determining whether voluntary
remuneration committees align CEO remuneration with ?rm ?nancial performance.
Companies having remuneration committees comprising all NEDIRs do not always
align CEO total pay or CEO bonuses with ?rm ?nancial performance. Our study ?nds
that large ?rm remuneration committees align both CEO total remuneration and CEO
bonuses with ?rm ?nancial performance at all levels of remuneration committee
independence; although, the most effective committees (i.e. with model coef?cients
Variable b SE t p-value
Intercept 8.019 1.025 7.820 0.000
Energy and mining 20.510 1.357 20.376 0.708
Industrials 22.613 1.490 21.754 0.082
Consumer 22.983 1.246 22.394 0.018
Services 0
a
[Size ¼ 0 smaller]
*
[REMCindep ¼ 0]
*
ROAlag1 20.016 0.078 20.202 0.840
[Size ¼ 0 smaller]
*
[REMCindep ¼ 1]
*
ROAlag1 0.067 0.154 0.434 0.665
[Size ¼ 0 smaller]
*
[REMCindep ¼ 2]
*
ROAlag1 0.022 0.092 0.243 0.808
[Size ¼ 1 med]
*
[REMCindep ¼ 0]
*
ROAlag1 0.042 0.220 0.190 0.849
[Size ¼ 1 med]
*
[REMCindep ¼ 1]
*
ROAlag1 0.187 0.184 1.016 0.311
[Size ¼ 1 med]
*
[REMCindep ¼ 2]
*
ROAlag1 0.423 0.180 2.345 0.021
* *
[Size ¼ 2 large]
*
[REMCindep ¼ 0]
*
ROAlag1 1.015 0.374 2.718 0.007
* * *
[Size ¼ 2 large]
*
[REMCindep ¼ 1]
*
ROAlag1 0.359 0.184 1.948 0.054
* *
[Size ¼ 2 large]
*
[REMCindep ¼ 2]
*
ROAlag1 0.523 0.166 3.158 0.002
* * *
Notes: Signi?cance at:
*
p , 0.10,
* *
p , 0.05,
* * *
p , 0.01 and
* * * *
p , 0.001 (all p-values are two-
tailed);
a
this parameter is set to zero because it is redundant:
CEObonus ¼ b0 þ b1 ðSize
*
ROAlag1
*
REMCindepÞ þ b2 Industry þ e
CEObonus – natural logarithm of CEO bonuses; Size – total assets ($ value) categorised into thirds: 0
– smaller ?rms, 1 – medium ?rms, 2 – larger ?rms; ROAlag1 – return on assets for the prior year
2000; REMCindep is the proportion of NEDIRs on the remuneration committee (RC), categorised into 0
– ?rms with no RCs or a minority NEDIRs, the least independent RCs; 1 – a majority of NEDIRs on
?rm RCs but less than 100 per cent; 2 – 100 per cent NEDIR composition, the most independent RCs;
Industry – industry groups based on GIC sectors, 1 – energy and mining, 2 – industrials, 3 –
consumer and 4 – services
Table XIII.
Model 6 CEO total
bonuses, parameter
estimates for three-way
interaction between ?rm
size remuneration
committee independence
and ?rm ?nancial
performance
Remuneration
committee
independence
215
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
having the smallest p-values) have 100 per cent NEDIR membership. The study also
shows that smaller ?rmremuneration committees are ineffective in reducing con?icts of
interest in relation to CEO total remuneration and bonuses for ?rm ?nancial
performance regardless of the level of their remuneration committee independence.
Results for medium-sized ?rms are mixed. Those medium-sized ?rms with
remuneration committees with a majority of NEDIRs link CEO total remuneration
with ?rm ?nancial performance, while medium-sized ?rms with remuneration
committees comprising 100 per cent NEDIRs link CEO bonuses with ?rm ?nancial
performance.
Our evidence suggests that the Australian Government Productivity Commission’s
(2009) recommendation to mandate independent remuneration committees to align
CEO remuneration with ?rm ?nancial performance may not be effective public policy,
particularly for medium and smaller ASX300 ?rms. To address corporate CEO
remuneration excesses, policy makers need to be aware of additional factors, such as
remuneration committee quality (Sun et al., 2009), because larger ?rms have the
resources to attract and employ higher quality human and intellectual resources
including quality independent directors (Australian Government Productivity
Commission, 2009). Moreover, the CEOs of large ?rms are more likely to
achieve appropriate ?rm ?nancial performance supported by extensive resources
(Merhebi et al., 2006).
Our study shows that CEOs of smaller ?rms receive full remuneration (including
bonuses), but that this remuneration has weak links to performance. This ?nding
suggests that the CEOs of smaller ?rms appear to attract less public and market scrutiny
than their larger counterparts and that smaller ?rms pay their CEOs regardless of
whether these ?rms perform or not. In the US regulatory context, where independent
compensation committees have been mandatory since 2003, Sun et al. (2009) found that
higher compensation committee quality, comprising a multi-dimensional measure, led to
a greater incentive alignment in executive compensation contracts in relation to future
?rm ?nancial performance. Future research could compare the 2001 sample of ASX 300
?rms with a sample of 2012 ASX 300 ?rms to see the regulatory impact of increased
remuneration committee independence. Additionally, the quality of remuneration
committee composition in the more regulated Australian context could also be examined.
While our study provides evidence about remuneration committee independence
interventions in ASX300 companies, there are limitations. Our sample of ?rms was
restricted to 2001, a time when it was possible to capture varying levels of remuneration
committee independence prior to increased corporate governance regulations. As with
most social science research, the models used to test the hypotheses are not complete and
we acknowledge that other important factors or interaction terms may not have been
included when investigating the relationship between CEO total pay and ?rm ?nancial
performance. Multicollinearity, however, becomes a problem if too many ?nancial
variables are included in the same model. We also acknowledge that this study is
unlikely to be completely free of endogeniety because “the potential for endogeniety
exists for virtually all studies involving accounting, ?nance and economics variables”
(Chenhall and Moers, 2007, p. 177). As they suggest, we have attempted to deal with
potential endogeniety by identifying the dependent variable and
independent/explanatory variables to be included in the models based on theory and
prior empirical evidence.
ARJ
26,3
216
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Notes
1. The Corporate Law Economic Reform Program (Audit Reform and Corporate Disclosure)
Act (2004) included s 250R(2) in the Corporations Act 2001 (Cth) as part of the legislative
reforms to the disclosure of director and executive remuneration. “Section 250R(2) requires
that a resolution that the remuneration report (attached to the directors’ report) be adopted
be put to shareholder vote at the annual general meeting (AGM) of listed companies. Section
250R(3) provides that this vote is to be non-binding. While a majority of voting shareholders
may vote against the proposed remuneration, the ultimate decision whether to accept the
report rests with the board” (Chapple and Christensen, 2005, p. 263). Shareholders’
non-binding vote on remuneration appears to have resulted in improved communication
between boards and shareholders and enhanced remuneration practices, despite initial
opposition by the local business community (Fels, 2010). The problem is that some boards
have ignored shareholders’ non-binding vote regarding executive remuneration (Fels, 2010).
In fact, the Australian Corporations Act was amended in 2011 to enact shareholders’ binding
vote or the Two Strike Rule. When there are two successive negative votes on board
remuneration reports at annual general meetings by 25 per cent or more of shareholders, the
company must formally respond by asking all board members except the CEO to stand for
re-election within 90 days. One-third of all board positions in Australian public companies
are elected every year (Sheehan, 2012).
2. The Productivity Commission undertook a public inquiry into the regulatory framework
around the remuneration of directors and executives of companies regulated under the
Corporations Act. The report was presented to the Australian Government on 19 December
2009 and released publicly on 4 January 2010. Speci?cally, the Commission was requested to
consider the following: the effectiveness of the existing framework for the oversight,
accountability and transparency of remuneration practices in Australia including; the role,
structure and content of remuneration disclosure and reporting; the scope of who should be
the subject of remuneration disclosure and; approving remuneration packages; the role of
boards and board committees in developing and approving; remuneration packages
(Australian Government Productivity Commission, 2009, p. v).
3. “The S&P/ASX 300 provides up to an additional 100 small cap stocks to the S&P/ASX200.
Index constituents are drawn fromeligible companies listed on the Australian Stock Exchange.
This index is designed to address investment managers’ needs to benchmark against a
portfolio characterized by suf?cient size and liquidity. [. . .]” (www.standardandpoors.com/
indices/sp-asx).
References
AASB (2004), Accounting Standard AASB 1046 Director and Executive Disclosures by Disclosing
Entities, Australian Accounting Standards Board, Melbourne.
AASB (2005), Accounting Standard AASB 101 Presentation of Financial Statements, Australian
Government Australian Accounting Standards Board, Melbourne.
AASB (1999), Accounting Standard AASB 1034 Financial Report Presentation and Disclosures,
Australian Accounting Standards Board, Australian Accounting Research Foundation,
Melbourne.
Australian Government Productivity Commission (2009), Executive Remuneration in Australia,
Report No. 49, Final Inquiry Report, Productivity Commission, Melbourne, available at:
www.pc.gov.au
Australian Securities Exchange (2007), Corporate Governance Principles and Recommendations with
2010 Amendments, 2nd ed., ASX Corporate Governance Council, Sydney, available at: www.
asxgroup.com.au/media/PDFs/cg_principles_recommendations_with_2010_amendments.pdf
Remuneration
committee
independence
217
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Australian Securities Exchange (2010a), Full List of Current GICS Sectors, available at: www.asx.
com.au/products/sector-indices.htm#consumer_discretionary_index
Australian Securities Exchange (2010b), “What is GICS?”, available at: www.asx.com.au/
products/gics.htm
Australian Securities and Investments Commission (1998), Practice Note 68: New Financial
Reporting and Procedural Requirements, Australian Securities and Investments
Commission, available at: www.asic.gov.au/asic/pd?ib.nsf/LookupByFileName/pn68.pdf/
$?le/pn68.pdf
Australian Stock Exchange (2003), Principles of Good Corporate Governance and Best
Practice Recommendations, ASX Corporate Governance Council, Sydney, March,
available at: www.asx.com.au/documents/about/principles-and-recommendations-march-
2003.pdf
Baber, W.R., Janakiraman, S.N. and Kang, S. (1996), “Investment opportunities and the
structure of executive compensation”, Journal of Accounting & Economics, Vol. 21 No. 3,
pp. 297-318.
Baker, G.P., Jensen, M.C. and Murphy, K.J. (1988), “Compensation and incentives: practice vs
theory”, The Journal of Finance, Vol. 43 No. 3, pp. 593-616.
Beatty, R.P. and Zajac, E.J. (1994), “Managerial incentives, monitoring, and risk bearing: a study
of executive compensation, ownership, and board structure in initial public offerings”,
Administrative Science Quarterly, Vol. 39 No. 2, pp. 313-335.
Berle, A.A. Jr and Means, C.G. (1932), The Modern Corporation and Private Property,
The Macmillan Company, New York, NY.
Black, F. and Scholes, M. (1973), “The pricing of options and corporate liabilities”, Journal of
Political Economy, Vol. 81 No. 3, pp. 637-654.
Boyer, R. (2005), “From shareholder value to CEO power: the paradox of the 1990s”, Competition
& Change, Vol. 9 No. 1, pp. 7-47.
Brambles (2001), Brambles Industries 2001 Annual Report, Brambles Industries PLC,
available at: www.brambles.com/BXB/Company/ShowPage.aspx/PDFs/3253-10000000/
2001AnnualReport156KB
Brunello, G., Graziano, C. and Parigi, B. (2001), “Executive compensation and ?rm
performance in Italy”, International Journal of Industrial Organization, Vol. 19 Nos 1/2,
pp. 133-161.
Cadbury, A. (1992), Report of the Committee on the Financial Aspects of Corporate Governance,
Gee, London.
Capezio, A., Shields, J. and O’Donnell, M. (2011), “Too good to be true: board structural
independence as a moderator of CEO pay-for-?rm-performance”, Journal of Management
Studies, Vol. 48 No. 3, pp. 487-513.
Carpenter, M.A. and Sanders, W.G. (2002), “Top management team compensation: the missing
link between CEO remuneration and ?rm performance?”, Strategic Management Journal,
Vol. 23 No. 4, pp. 367-375.
Chapple, E. and Christensen, B. (2005), “The non-binding vote on executive pay: a review of the
CLERP 9 reform”, Australian Journal of Corporate Law, Vol. 18, pp. 263-287.
Chenhall, R.H. and Moers, F. (2007), “The issue of endogeniety in theory-based,
quantitative management accounting research”, European Accounting Review, Vol. 16 No. 1,
pp. 173-195.
ARJ
26,3
218
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Clarkson, P., Walker, J.K. and Nicholls, S. (2011), “Disclosure, shareholder oversight
and pay-to-performance link”, Journal of Contemporary Accounting and Economics, Vol. 7,
pp. 47-64.
Conyon, M.J. and Peck, S.I. (1998), “Board control, remuneration committees, and
top management compensation”, Academy of Management Journal, Vol. 41 No. 2,
pp. 146-157.
Daily, C.M., Dalton, D.R. and Cannella, A.A. Jr (2003), “Corporate governance: decades of
dialogue and data”, Academy of Management Review, Vol. 28 No. 3, pp. 371-382.
Daily, C.M., Johnson, J.L., Ellstrand, A.E. and Dalton, D.R. (1998), “Compensation committee
composition as a determinant of CEO compensation”, Academy of Management Journal,
Vol. 41 No. 2, pp. 209-220.
Dechow, P.M., Huson, M.R. and Sloan, R.G. (1994), “The effect of restructuring charges on
executives’ cash compensation”, The Accounting Review, Vol. 69 No. 1, pp. 138-156.
Ezzamel, M. and Watson, R. (1998), “Market comparison earnings and the bidding-up of
executive cash compensation: evidence from the United Kingdom”, Academy of
Management Journal, Vol. 41 No. 2, pp. 221-231.
Fama, E.F. (1980), “Agency problems and the theory of the ?rm”, Journal of Political Economy,
Vol. 88 No. 2, pp. 288-307.
Fama, E.F. and Jensen, M.C. (1983), “Separation of ownership and control”, The Journal of Law
and Economics, Vol. 26 No. 2, pp. 301-325.
Fels, A. (2010), “Executive remuneration in Australia”, The Australian Accounting Review, Vol. 20
No. 1, pp. 76-82.
Friedrich, R.J. (1982), “In defense of multiplicative terms in multiple regression equations”,
American Journal of Political Science, Vol. 26 No. 4, pp. 797-833.
Gaver, J.J. and Gaver, K.M. (1998), “The relation between nonrecurring accounting
transactions and CEO cash compensation”, The Accounting Review, Vol. 73 No. 2,
pp. 235-253.
Godfrey, J., Mather, P. and Ramsay, A. (2003), “Earnings and impression management in
?nancial reports: the case of CEO changes”, Abacus, Vol. 39 No. 1, pp. 95-123.
Hagerman, R.L. and Zmijewski, M.E. (1979), “Some economic determinants of accounting
policy choice”, Journal of Accounting & Economics, Vol. 1 No. 2, pp. 141-161.
Hay Group (1998), Short-term Incentives Report 1998/99, Hay Group, Melbourne.
Hayes, A.F. (2005), Statistical Methods for Communication Science, Erlbaum, Mahwah, NJ.
Hayes, A.F. and Matthes, J. (2009), “Computational procedures for probing interactions in OLS
and logistic regression: SPSS and SAS implementations”, Behavior Research Methods,
Vol. 41 No. 3, pp. 924-936.
Hutchinson, M. andGul, F.A. (2004), “Investment opportunityset, corporate governancepractices and
?rm performance”, Journal of Corporate Finance, Vol. 10 No. 4, pp. 595-614.
Irwin, J.R. and McClelland, G.H. (2001), “Misleading heuristics and moderated multiple
regression models”, Journal of Marketing Research, Vol. 38 No. 1, pp. 100-109.
Izan, H.Y., Sidhu, B. and Taylor, S. (1998), “Does CEO remuneration re?ect performance?
Some Australian evidence”, Corporate Governance: An International Review, Vol. 6 No. 1,
pp. 39-47.
Jaccard, J. and Turrisi, R. (2003), Interaction Effects in Multiple Regression, 2nd ed., Quantitative
Applications in the Social Sciences No. 72, Sage, Thousand Oaks, CA.
Remuneration
committee
independence
219
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Jensen, M.C. (2000), A Theory of the Firm: Governance, Residual Claims, and Organizational
Forms, Harvard University Press, Cambridge, MA.
Jensen, M.C. and Meckling, W.H. (1976), “Theory of the ?rm: managerial behavior,
agency costs and ownership structure”, Journal of Financial Economics, Vol. 3 No. 4,
pp. 305-360.
Jensen, M.C. and Murphy, K.J. (1990), “Performance pay and top-management incentives”,
Journal of Political Economy, Vol. 98 No. 2, pp. 225-264.
Jensen, M.C., Murphy, K. and Wruck, E.G. (2004), “Remuneration: where we’ve been, how we
got to here, what are the problems, and how to ?x them”, Harvard NOM Working
Paper No. 04-28; ECGI – Finance Working Paper No. 44/2004, available at: http://ssrn.
com/abstract¼561305
Kent, P., Kercher, K. and Routledge, J. (2012), “Determinants of voluntary board
remuneration committee formation and composition”, paper presented at the
Accounting & Finance Association of Australia and New Zealand Conference,
Melbourne, Australia, 1-3 July 2012.
McCullagh, P. and Nelder, J.A. (1989), Generalized Linear Models, 2nd ed., Chapman & Hall,
London.
Merhebi, R., Pattenden, K., Swan, P.L. and Zhou, X. (2006), “Australian chief executive
of?cer remuneration: pay and performance”, Accounting & Finance, Vol. 46 No. 3,
pp. 481-497.
Nelder, J.A. and Wedderburn, R.W.M. (1972), “Generalized linear models”, Journal of the Royal
Statistical Society Series A (General), Vol. 135 No. 3, pp. 370-384.
Newman, H.A. and Mozes, H. (1999), “Does the composition of the compensation
committee in?uence CEO compensation practices?”, Financial Management, Vol. 28
No. 3, pp. 41-53.
Rainsbury, E.A., Bradbury, M. and Cahan, S.F. (2009), “The impact of audit committee quality on
?nancial reporting quality and audit fees”, Journal of Contemporary
Accounting & Economics, Vol. 5 No. 1, pp. 20-33.
Reynolds, J.K., Deis, D.R. Jr and Francis, J.R. (2004), “Professional service fees
and auditor objectivity”, Auditing: A Journal of Practice and Theory, Vol. 23 No. 1,
pp. 29-52.
Sarbanes-Oxley Act (2002), available at: www.pcaobus.org/ (accessed 17 July 2010).
Shaw, K.W. and Zhang, M.H. (2010), “Is CEO cash compensation punished for poor ?rm
performance?”, The Accounting Review, Vol. 85 No. 3, pp. 1065-1093.
Sheehan, K. (2009), “The regulatory framework for executive remuneration in Australia”,
Sydney Law Review, Vol. 31 No. 2, pp. 273-308.
Sheehan, K. (2012), The Regulation of Executive Compensation: Greed, Accountability and
Say on Pay, Edward Elgar Publishing Limited, Cheltenham.
Skinner, D.J. (1994), “The investment opportunity set and accounting procedure
choice: preliminary evidence”, Journal of Accounting & Economics, Vol. 16 No. 4,
pp. 407-445.
Stiglitz, J.E. (2010), Freefall: America, Free Markets, and the Sinking of the World Economy,
W.W. Norton & Co, New York, NY.
Sun, J., Cahan, S.F. and Emanuel, D. (2009), “Compensation committee governance quality, chief
executive of?cer stock option grants, and future ?rm performance”, Journal of
Banking & Finance, Vol. 33 No. 8, pp. 1507-1519.
ARJ
26,3
220
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The Corporate Law Economic Reform Program (Audit Reform and Corporate Disclosure)
Act (2004), available at: www.asic.gov.au (accessed 17 July 2010).
Tosi, H.L. Jr, Werner, S., Katz, J.P. and Gomez-Mejia, L.R. (2000), “How much does performance
matter? A meta-analysis of CEO remuneration studies”, Journal of Management, Vol. 26
No. 2, pp. 301-339.
Vafeas, N. (2003), “Further evidence on compensation committee composition as a determinant of
CEO compensation”, Financial Management, Vol. 32 No. 2, pp. 53-70.
Corresponding author
Carolyn Windsor can be contacted at: [email protected]
Remuneration
committee
independence
221
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Michael Seamer, Adrian Melia. 2015. Remunerating non-executive directors with stock options: who is
ignoring the regulator?. Accounting Research Journal 28:3, 251-267. [Abstract] [Full Text] [PDF]
D
o
w
n
l
o
a
d
e
d
b
y
P
O
N
D
I
C
H
E
R
R
Y
U
N
I
V
E
R
S
I
T
Y
A
t
2
1
:
1
8
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_311141178.pdf