Description
The purpose of this paper is to test for managerial opportunism, specifically the
backdating of executive options, in Australia.
Accounting Research Journal
Evidence of managerial opportunism in Australia
Andrew Trumble Sean Pinder
Article information:
To cite this document:
Andrew Trumble Sean Pinder, (2012),"Evidence of managerial opportunism in Australia", Accounting
Research J ournal, Vol. 25 Iss 1 pp. 25 - 40
Permanent link to this document:http://dx.doi.org/10.1108/10309611211244492
Downloaded on: 24 January 2016, At: 21:14 (PT)
References: this document contains references to 10 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 2635 times since 2012*
Users who downloaded this article also downloaded:
Kim Ittonen, (2012),"Market reactions to qualified audit reports: research approaches", Accounting
Research J ournal, Vol. 25 Iss 1 pp. 8-24http://dx.doi.org/10.1108/10309611211244483
Irene Tempone, Marie Kavanagh, Naomi Segal, Phil Hancock, Bryan Howieson, J enny Kent,
(2012),"Desirable generic attributes for accounting graduates into the twenty-first century:
The views of employers", Accounting Research J ournal, Vol. 25 Iss 1 pp. 41-55 http://
dx.doi.org/10.1108/10309611211244519
Ismail Cebeci, (2012),"Integrating the social maslaha into Islamic finance", Accounting Research J ournal,
Vol. 25 Iss 3 pp. 166-184http://dx.doi.org/10.1108/10309611211290158
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Evidence of managerial
opportunism in Australia
Andrew Trumble and Sean Pinder
Department of Finance, University of Melbourne,
Parkville, Australia
Abstract
Purpose – The purpose of this paper is to test for managerial opportunism, speci?cally the
backdating of executive options, in Australia.
Design/methodology/approach – The paper analyses the return behaviour associated with a
sample of 161 unscheduled options granted by Australian ?rms. Speci?cally, the authors test for
differences between a subsample of grants that had late-?led notices (and hence may be subject to
backdating) versus those that had notices ?led on-time.
Findings – Consistent with backdating, it is found that these abnormal post-grant returns persist for
a sub-sample of late-?led grants but not for a sub-sample of grants with same-day ?ling. Furthermore
– the authors ?nd even stronger results for option grants made by ?rms with a history of late-?ling
but for which no notice was ?led with the Australian Securities Exchange. This paper is the ?rst to
demonstrate these effects in a setting subject to the IFRS requirement that the fair value (rather than
the intrinsic value) of executive options be expensed.
Originality/value – This paper is the ?rst to demonstrate these effects in Australia and further in a
setting subject to the IFRS requirement that the fair value (rather than the intrinsic value) of executive
options be expensed.
Keywords Options backdating, Managerial opportunism, Executive compensation, Australia, Pay
Paper type Research paper
1. Introduction
As a form of performance-sensitive remuneration, executive options have become a
popular means of aligning the interests of rent-extracting executives and pro?t-seeking
shareholders. A stream of US-based literature, however, details how executive option
grants have been used to covertly endow their recipients with performance-insensitive
pay. Yermack (1997) documents positive average abnormal returns over the period of
time immediately following the grant dates of CEO options. He interprets this as
evidence that executives “spring-load” their option grants by opportunistically timing
them to precede foreseeable stock price appreciations. Lie (2005) observes a negative
average abnormal return over the period of time preceding the grant dates of CEO
options and a positive average abnormal return immediately following these dates.
He contends that these abnormal returns are likely attributable to undisclosed “options
backdating”.
Options backdating refers to the practice by which a ?rm falsely claims to have
granted executive options on a past date. The past date nominated will be a date on
which the ?rm’s stock price was lower than the date on which the ?rm makes the false
claim (the current date). This way, secretly backdated options, which appear to have
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The authors would like to thank Rob Brown, John Handley, Steve Easton and David Yermack for
their helpful comments and advice on previous versions of this paper.
Managerial
opportunism
in Australia
25
Accounting Research Journal
Vol. 25 No. 1, 2012
pp. 25-40
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611211244492
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
been endowed with a given amount of intrinsic value on the past date, assume
greater intrinsic value as of the current date. Undisclosed backdating of this nature
thereby allows ?rms to understate the true intrinsic value of their executives’ option
grants.
In 2006, backdating became front-page news following a report in The Wall Street
Journal that the practice was endemic within corporate America (Forell and Bandler,
2006). The Securities and Exchange Commission (SEC) and the Department of Justice
have since come to view the practice as fraudulent and in a number of well publicized
cases both civil and criminal charges have been laid[1]. Despite the pro?le of the
backdating problem in the USA, there is no published evidence that the practice also
exists in Australia. Under the Australian Securities Exchange (ASX) Listing Rules
relating to continuous disclosure, ?rms must ?le notice of option grants “at the same
time” as they are granted (ASX Listing Rule 3.10.5), which should make backdating
impracticable. That is, the options will always be reported as having been granted on a
past date when ?ling a grant of backdated options. As such, all backdated options would
have the appearance of being ?led some time after they were reportedly granted and
thus would appear to breach Listing Rule 3.10.5. Notably however, Heron and Lie (2007)
demonstrated that following the implementation of the Sarbanes Oxley Act (SOX) in
2002, US ?rms consistently de?ed the SEC requirement of ?ling notice of options grants
within two days of their granting. The present study therefore investigates whether
Australian ?rms have a similar record of late-?ling which, if demonstrated, implies that
backdating in Australia is a distinct possibility.
We gather a sample of 161 unscheduled stock options granted to the key executives
(including but not limited to the CEO) of 61 ASX200 companies between January 1, 2005
and July 30, 2007. Consistent with some form of opportunism, be it spring-loading or
backdating, we observe a positive average abnormal return of 2.02 percent in the 25-day
post-grant window. A test to determine if this abnormal return pattern is attributable
speci?cally to backdating is made worthwhile given that a number of the sampled
grants were discovered to have been late-?led. Categorising the sampled options
according to those which had same-day ?lings and those which were ?led a minimumof
two days late, we observe that a positive post-grant average abnormal return is retained
for the sub-sample of late-?led options, but disappears for the sub-sample of options ?led
on time. Where the grants in both of these sub-samples could have been spring-loaded,
but only the late-?led grants could have been backdated, these results are consistent
with backdating rather than spring-loading being the key form of opportunism
prevalent in the sample.
The remainder of the paper proceeds as follows. Section 2 provides an overview of
the extant literature which has focused on identifying irregularities in stock price
patterns surrounding the grant dates of executive options as well as an analysis of the
mechanics of options backdating, and details the motivations for backdating as
suggested by US research. This section also documents the differences between the US
and Australian disclosure requirements and accounting standards which pertain to the
practice of backdating. Section 3 details the test for evidence of general opportunism in
the sample of executive options grants, while Section 4 presents the direct test for
evidence of options backdating and then provides a test of the robustness of any
results relating to an increased sample of option grants. Section 5 outlines the
conclusions of the study.
ARJ
25,1
26
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Options backdating
2.1 Prior literature
The literature concerning stock return patterns surrounding executive option grants
remains almost exclusively US-based. Yermack (1997) provided the ?rst evidence that
option grants were providing a means for rent extraction. Sampling Fortune
500 companies from 1992 to 1994, Yermack found that ?rms making stock option
awards to their CEOs outperformed the market on a risk-adjusted basis by slightly
more than 2 percent during the 50-day post-grant period. Yermack saw this result as
consistent with his hypothesis that CEOs were setting the terms of their own
remuneration packages. Speci?cally, Yermack suggested that executives were having
their option grants timed to be awarded just prior to announcements of favourable
company news, a practice which has become known as spring-loading. Yermack
further substantiated his claim that the observed abnormal return pattern re?ected
undue managerial in?uence by demonstrating that where a granting ?rm’s CEO sat on
its remuneration committee, the 50-day post-grant average abnormal return rose to
11.2 percent.
Aboody and Kasznik (2000) noted the way in which many ?rms scheduled
compensation committee meetings – at which executive options were known to be
granted – to approximately the same date each year. Where these meetings were
scheduled up to a year in advance, the authors noted that options granted at the
meetings could not have been spring-loaded. Accordingly, in analyzing return patterns
surrounding CEO option grants made between 1992 and 1996, Aboody and Kasznik
categorised options as either “scheduled” or “unscheduled” to distinguish between
options which appeared to have been granted on a ?xed annual basis at scheduled
compensation committee meetings, and options which did not. Speci?cally, the study
classi?ed scheduled options as those which had grant dates within one week of the one
year anniversary of a previous grant made by the same ?rm. For the unscheduled
sub-sample, where spring-loading was still thought possible, the study not only found
a positive mean abnormal return in the 30-day post-grant window, but also a negative
mean abnormal return in the 30-day pre-grant window. The study also revealed a
positive mean abnormal return in the 30-day post-grant window for the sub-sample of
scheduled grants. This result led the authors to conclude that while CEOs may not
have been able to opportunistically time the grant dates of scheduled options, they
were still able to in?uence the payoff on these options by leaking price sensitive
information on the days surrounding their grant dates.
Lie (2005) was the ?rst study to empirically identify the practice of options
backdating. Partitioning option grants made between 1992 and 2002 into scheduled
and unscheduled samples (using the same criteria as Aboody and Kasznik (2000)),
Lie found that while both samples exhibited a v-shaped pattern of negative pre-grant
and positive post-grant average abnormal returns, the pattern for the sample of
unscheduled grants was far more pronounced. Lie noted that the price movements
surrounding the dates of the unscheduled option grants were being driven by
market-wide factors which managers were unlikely to have been predicting. This
meant that it was unlikely that spring-loading was behind the abnormal pattern of
returns. As a means of explaining the return patterns, Lie instead suggested that in
order to secretly endow executive options with hidden amounts of intrinsic value, ?rms
may have been retroactively declaring option grants to have been made on past dates
Managerial
opportunism
in Australia
27
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
on which their company stock price was low. In that this “backdating” could not
have been effected on options with grant dates that were thought to be ?xed in
advance, this theory provided an elegant explanation as to why the study found
a stronger abnormal return pattern for the unscheduled sample than it did for the
scheduled sample[2].
Heron and Lie (2007) conducted further tests for evidence of backdating. Following
the implementation of Sarbanes Oxley, effective as of August 29, 2002, the acceptable
post-grant delay for ?ling – and thus the backdating window for any ?rm that wanted
to ?le on time – was reduced signi?cantly to just two business days. Heron and
Lie hypothesized that if backdating had caused the v-shaped pattern of abnormal
returns observed for the sample of unscheduled pre-SOX grants in Lie (2005), then
any abnormal return pattern observable for a sample of post-SOX grants would
likely be much weaker. To test their hypothesis, Heron and Lie gathered two samples
of unscheduled options, a post-SOX sample spanning from August 29, 2002 to
November 30, 2004 and a sub-sample of the pre-SOX grants sampled in Lie (2005)[3].
They found that “the magnitude of the average abnormal return during the week
before (after) the grants (was) roughly six (?ve) times larger” for the pre-SOX sample
than it was for the post-SOX sample (Heron and Lie, 2007, p. 273).
Heron and Lie then sought to con?rm that this weaker but still signi?cant post-SOX
return pattern was attributable to backdating rather than perhaps spring-loading. The
post-SOXoptions were grouped according to the number of days lag between their grant
dates and ?ling dates. Heron and Lie conjectured that if backdating had continued
post-SOX, the abnormal return patterns for the sub-samples would become more
pronounced as the lag increased and the backdating windowwidened. The study found
that a sub-sample of options with a one day ?ling lag exhibited no average abnormal
returns on the days surrounding the grants, but that a sub-sample of grants with a
two-daylagexhibiteda positive dayone average abnormal return. Further, a sub-sample
with a three-day lag exhibited positive average abnormal returns on both day one and
day two, while a sub-sample of grants with a lag of four or more days exhibited a total
re-emergence of the v-shaped pattern of average abnormal returns in all of the event
windows examined. These results not only suggested that US ?rms, post-SOX, deemed
it worthwhile to backdate over windows as small as two days, but also that they were
prepared to breach the two-day ?ling requirement in order to backdate over longer
periods.
2.2 The mechanics of, and motivation for, options backdating
Options backdating refers to the practice by which a ?rm falsely claims to have
granted executive options on a past date. The past date nominated will be a date on
which the ?rm’s stock price was lower than the date on which the ?rm makes the false
claim (the current date). In this situation, secretly backdated options said to have been
granted at-the-money (for example) on the nominated past date will accordingly be
in-the-money as at the current date. Where a ?rm keeps such backdating secret, it will
appear to outsiders (shareholders and regulators included) that the ?rm granted
options with a lower intrinsic value than their actual worth.
Listed ?rms in both the USAand Australia are required to submit ?lings of executive
option grants to their securities exchanges within a set timeframe following the grant
date. Presently, US ?rms have two business days, while Australian ?rms are expected
ARJ
25,1
28
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
theoretically to ?le on the day of the grant[4]. Because backdating involves nominating a
past date as the date on which an option was said to have been granted, there will always
be a lag between the nominated grant date of a backdated option and the date on which
the option is ultimately ?led. Therefore, if a ?rm wishes to backdate an option but also
wants to be seen to be ?lingontime, the windowover whichthe ?rmcanbackdate is only
as long as the permissible lag in ?ling. For example, consider the US ?rm has a two-day
window in which to ?le options following their granting. If the ?rm secretly backdated
an option by two days and then ?led immediately, the exchange would be left to believe
that the ?rm had physically granted the option two days earlier and had simply waited
its full measure before ultimately ?ling on time. By contrast, as Australian ?rms are
required to ?le notice of options immediately upon their granting, it would be impossible
to both backdate options and meet the ?ling requirement. Australian backdaters
therefore could not avoid appearing as late-?lers.
Prior to June 15, 2005, the accounting treatment for executive stock option grants in
the USA was governed by Accounting Principles Board (APB) Opinion Number
25 Accounting for Stock Issued to Employees. Opinion 25 required US ?rms to expense
the grant date intrinsic value of any executive options granted; options secretly
backdated to appear at-the-money therefore carried no compensation charge for the
?rm. Where these backdated at-the-money options would have intrinsic value as of
their real grant dates, substituting them for other forms of executive compensation
would have provided a means for US ?rms to reduce their remuneration expense and
boost reported pro?t without necessarily reducing the effective level of payout to their
executives. Had the US ?rms that backdated done so for this purpose, it would have
been expected that, relative to their peers, they would have paid lower levels of other
forms of executive compensation. However, in what is a telling result, Bebchuk et al.
(2010) found that relative to their peers, US ?rms suspected of backdating over the
period 1996-2005 were in fact paying higher levels of executive compensation in other
forms.
Given that backdated options therefore appear to be adding to, rather than
substituting for, other forms of executive remuneration, undisclosed backdating has
come to be seen as a means of rent extraction symptomatic of executives exerting
excessive in?uence over their own remuneration. Consistent with this managerial
power theory, a number of papers (Bizjak et al., 2009; Collins et al., 2009; Bebchuk et al.,
2010) demonstrate that a variety of ?rm characteristics conducive to excessive CEO
power and indicative of poor governance were signi?cant in determining which ?rms
appeared to be backdating their grants. For example, Collins et al. (2009) showed that
the likelihood of backdating increased with the percentage of inside directors on the
board, the presence of inside directors on the audit committee, the presence of outside
directors hired by the incumbent CEO, and where the CEO doubled as chairperson.
In what is a particularly striking result, the study also showed that a CEO’s grant was
more likely to have been backdated where an additional grant was made to a company
director on the same-day.
2.3 Backdating in Australia
Our sample consists of executive options granted in Australia after January 1,
2005. There are important differences between this Australian environment and
the pre-2005 US environment that has been the focus of extant studies into
Managerial
opportunism
in Australia
29
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
options backdating. We demonstrate that not only do these differences not necessarily
preclude the possibility of backdating in our sample but also that our sampled ?rms
and their executives remain in a position to realise large ?nancial bene?ts as a result
of options backdating.
2.3.1 Australian ?ling requirements. Under the ASX continuous disclosure
requirements, ASX Listing Rule 3.10.5 stipulates that listed Australian ?rms, through
the ?ling of an “Appendix 3B: new issue announcement: application for quotation of
additional securities and agreement” (Appendix 3B) notice, are required to notify the
ASX of option grants “at the same time” as they are granted. Notably, because
backdaters reporting their grants choose to cite a past date as the grant date, any
backdated options in Australia would appear to be late-?led and in breach of Listing
Rule 3.10.5. Our investigation into Australian backdating would therefore have been
made redundant if ?rms strictly adhered to the requirement. Importantly, though,
it appears they do not. Speci?cally, as we report in Section 3 of the paper, we observe
that an Appendix 3B was ?led for only approximately 44 percent of the sampled option
grants and that only about a third of these Appendix 3Bs were ?led on the same date
that the options were reported to have been granted.
2.3.2 Australian accounting standards. The expensing requirement for employee
options determines howconcealing an option’s intrinsic value via backdating might also
result in ?rms under-reporting the appropriate level of remuneration expense.
Fundamentally, the requirement will not alter a backdaters ability to conceal intrinsic
value – and to this end will not in?uence the primary incentive to backdate.
The expensing requirement for Australian ?rms in our sample is slightly different from
the requirement imposed upon the US ?rms sampled in prior studies. Speci?cally, prior
to June 15, 2005, APB Opinion Number 25, Accounting for Stock Issued to Employees
required US ?rms to expense the grant date intrinsic value of their executive options.
In contrast, our post January 1, 2005 sample of ?rms were required to expense the grant
date fair value of their executive options, as per AASB 2: Share-based Payment[5].
Despite this slight difference, because backdating involves understating the actual
intrinsic value and thus also the actual fair value of options, both US and Australian
backdaters would have reported lower compensation expense and boosted their
reported pro?tability, while at the same time concealing from the market the true value
of the compensation package provided to their executives.
3. Test for evidence of general opportunism
3.1 Sample
We sample options granted to company executives – including but not limited to the
CEO – of ?rms in the ASX200 as at June 30, 2007. As we only include option grants that
were subject to International Financial Reporting Standards (IFRS), the starting date for
the sample is January 1, 2005.
The sampled options, speci?cally their grant dates and exercise prices, are taken
from company end-of-?nancial-year reports. If a company made multiple grants on the
same date with different exercise prices, we sample only the grant with the lowest
exercise price. If a company made multiple grants on the same date with the same
exercise price (but perhaps with different vesting schedules) we randomly sample only
one of these grants.
ARJ
25,1
30
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In the USA, the compensation committee meeting at which a ?rm’s executive
options are granted is typically scheduled to fall on approximately the same date each
year. Because the grant dates of options granted at these meetings are ?xed, such
“scheduled” options are not deemed susceptible to spring-loading or backdating[6]. For
the Australian ?rms whose annual reports we survey, a number of their reported
options similarly appear to have been granted on an annual schedule and are thus
excluded from our investigation into opportunism. Speci?cally, we deem an option to
have been scheduled, and thus omit it from the sample, if we observe that another
grant has been made by the same ?rm within one day of the anniversary of the option,
either in the year before, or in the year after the option was granted[7].
Table I displays the key summary statistics for the ?nal sample. The sample
comprises 161 unscheduled executive option grants, made by 61 different companies.
The intrinsic value of the options is determined according to the closing stock price on
the reported grant date.
3.2 Methodology
We measure abnormal returns on a continuously compounding basis over a 51-day
event window surrounding the grant dates of the 161 sampled unscheduled options.
Normal returns on a stock are estimated according to a market model comprising stock
betas which were estimated over a ?ve-month estimation window ending
approximately four months prior to the grant date[8]. From Datastream we obtained
total returns data on the value-weighted S&P ASX 200 Index which was taken as a
proxy for the market portfolio. Individual stock return data and standard proxies for
risk-free rates of return were obtained from the IRESS database.
The test for evidence of general opportunism is a general test in that any positive
abnormal pattern in returns could stem from either spring-loading, backdating or both.
Given that the fundamental aim of both backdating and spring-loading is to capture
post-grant stock price appreciations, a failure to observe positive post-grant abnormal
returns would be inconsistent with both practices. In terms of the pre-grant period,
there is no return pattern that would be inconsistent with either practice. It should
be noted that two prominent US studies of backdating – Lie (2005) and Heron and
Lie (2007) – both documented negative average abnormal returns in the pre-grant
period. These results indicate that in striving for greater gains, US ?rms backdated
options to dates at which their company stock price was at a local minimum. Indeed,
Heron and Lie (2007) demonstrated that post-SOX, US ?rms still appeared to backdate
General
Total number of unscheduled grants 161
Total number of ?rms 61
Intrinsic value of options granted
Options granted were out-of-the-money 62
Options granted were at-the-money 10
Options granted were in-the-money 89
Notes: Descriptive statistics for a sample of unscheduled executive stock option grants made by
ASX200 ?rms (according to the composition at June 30, 2007) between January 1, 2005 and June 30,
2007; the intrinsic value of an option was determined according to the closing stock price on the
reported grant date of the option
Table I.
Descriptive statistics for
sample of unscheduled
executive options grants
Managerial
opportunism
in Australia
31
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to the local minimum even when this meant backdating by considerably more than two
days, and accordingly breaching the two-day ?ling deadline by considerable margins[9].
This local minimum-at-all-costs approach is not however a necessity when backdating.
Had the US backdaters seen greater risk of detection in ?ling excessively late, then they
may well have elected to stop short of local minimums, such that the pre-grant abnormal
returns might have appeared positive. Ultimately this study will report pre-grant
abnormal returns so as to aid comparability with the prior US studies; however, in
identifying evidence of opportunism, our focus will be on post-grant abnormal returns.
3.3 Results
Figure 1 and Table II display the average abnormal returns over the 51-day window
surrounding the grant dates for the sample of 161 unscheduled option grants.
The positive average abnormal return of 2.02 percent in the 25-day post-grant
period is consistent with the presence of some form(s) of opportunism, be it backdating,
spring-loading or both.
Figure 1.
Cumulative average
abnormal returns
surrounding the grant
date of executive stock
option grants
4.00%
3.50%
3.00%
2.50%
2.00%
1.50%
0.50%
0.00%
–25–23–21–19–17–15–13–11 –9 –7 –5 –3 –1 25 23 21 19 17 15 13 11 9 7 5 3 1
Trading days relative to reported grant date
C
u
m
u
l
a
t
i
v
e
A
v
e
r
a
g
e
A
b
n
o
r
m
a
l
R
e
t
u
r
n
1.00%
All option grants
Notes: This figure shows the cumulative average abnormal stock returns from 25 days prior
to 25 days after the grant date for a sample of unscheduled executive stock option grants made
by ASX 200 companies between January 1, 2005 and June 30, 2007; abnormal returns are
estimated using a market model of normal returns based on a five month estimation period
ending four months prior to the grant date
ARJ
25,1
32
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4. Direct test for evidence of options backdating
4.1 Sample
We are interested in determining whether the average abnormal post-grant return
observed for the sample of executive option grants is attributable to backdating.
Where only late-?led grants can have been backdated, the possibility that backdating
was behind the average abnormal return can only be considered if there is evidence of
late-?ling in the sample. To determine if there is, we search the ASX web site www.asx.
com.au where ?lings of Appendix 3Bs are announced and where copies of the
appendices are available. As shown in Table III, there is a high number of instances
(88 out of 161) where there appears to have been no announcement from the ASX of an
Appendix 3B having been ?led for options granted by a sample ?rm. This result
suggests that either the ASX systematically failed to notify the public of ?lings
received, or perhaps more seriously, that ?rms failed to submit Appendix 3Bs for these
particular options[10]. We therefore measure the ?ling lags for the remaining 73 option
grants for which a relevant Appendix 3B is identi?able. The lag between an option’s
grant date and its ?ling date is calculated according to the number of trading days
between the option’s grant date, as reported in the ?rm’s annual report[11], and the
date on which the ASX announced the option to have been ?led. Table III documents
the ?ling lags and the summary statistics for these 73 option grants.
The ?lings for 27 of these option grants were announced by the ASX two or more
trading days after the options’ respective grant dates. This result clearly suggests that
?rms are contravening the ASX requirement that executive options be ?led “at the
same time” as they are granted and accordingly makes backdating in Australia a
distinct possibility.
All grants
n ¼ 161
Window Mean p-value
25 20.0028 0.1058
24 0.0016 0.3493
23 0.0031 0.0610
*
22 0.0016 0.3120
21 0.0026 0.1212
0 0.0031 0.0992
*
1 0.0027 0.1213
2 0.0015 0.4296
3 0.0012 0.4756
4 0.0003 0.8889
5 20.0001 0.9680
(225,0) 0.0150 0.0883
*
(1,25) 0.0202 0.0305
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different from zero; this table presents daily average abnormal returns and
cumulative average abnormal returns surrounding unscheduled executive option grants made by ASX
200 companies between January 1, 2005 and June 30, 2007; abnormal stock returns are estimated using
a market model of normal returns based on a ?ve month estimation period ending four months prior to
the grant date; grant dates are taken from company end-of-year reports
Table II.
Abnormal stock returns
surrounding executive
options grants
Managerial
opportunism
in Australia
33
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4.2 Methodology
In testing directly for evidence of backdating, we create two sub-samples, one
comprising grants with same-day ?ling and the other comprising grants ?led two or
more days late[12]. If over the sample period Australian ?rms engaged in only options
backdating, the sub-sample of late-?led options ought to exhibit positive post-grant
abnormal returns, while the sub-sample of options ?led on time, which were not
backdated, should not. Alternatively, if spring-loading occurred then positive
post-grant abnormal returns ought to be observed in both sub-samples given that a
?rm’s choice to spring-load would be independent of its ?ling practices.
4.3 Results
Figure 2 and Table IVdemonstrate that there is a positive post-grant average abnormal
return for the sub-sample of late-?led options but not for the sub-sample of options with
same-day ?ling. Speci?cally, in the ?ve-week period following the grant, ?rms that ?led
grants two or more days late appear on average to have abnormal returns of almost
5 percent, while ?rms that ?led on time realised abnormal returns that, on average, were
not signi?cantly different from zero. This difference is consistent with backdating,
rather than spring-loading, being the key form of opportunism surrounding executive
stock option grants in Australia over the sample period[13].
Panel A: ?ling lags
Filing lag ,0 0 1 2 3 4 5 6 7 8 9 10 11 þ Subtotal
No ?ling
recorded Total
Frequency 2 25 19 4 8 2 1 1 2 1 1 0 7 73 88 161
Panel B: descriptive statistics
General
Total number of
unscheduled
grants 73
Total number of
?rms 33
Intrinsic value of
options granted
Options granted
were out-of-the-
money 29
Options granted
were at-the-
money 5
Options granted
were in-the-
money 39
Notes: Filing lags and descriptive statistics for a sub-sample of unscheduled executive stock option
grants made by ASX200 ?rms (according to the composition at June 30, 2007) between January 1, 2005
and June 30, 2007; the ?ling lag is the number of trading days between the grant date as reported in the
company end-of-year report and the date on which the ASX noti?ed the market via the company
announcements pages of the ASX web site: www.asx.com.au of an Appendix 3B ?led in relation to the
grant; the intrinsic value of an option was determined according to the closing stock price on the
reported grant date of the option
Table III.
Filing lags and
descriptive statistics for
the sample of
unscheduled executive
options grants
ARJ
25,1
34
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Having examined the late-?led options for evidence consistent with backdating,
it is important to also examine the 88 grants for which no Appendix 3B was observed to
have been lodged. As there were no apparent ?lings for these option grants, the ?rst
external noti?cation of them may well have come from the end-of-?nancial-year reports
from which we sampled. If this was the case, then these option grants, like the late-?led
ones, could also have been backdated. While an initial test of the 88-?led options
did not reveal any abnormal returns consistent with backdating, we noted that
25 of the non-?led options had been granted by ?rms that had also granted at least one of
the late-?led option grants. For these 25 options grants, Table V shows
evidence consistent with further options backdating. Speci?cally, the sub-sample of
these 25 grants, like the late-?led sub-sample, also exhibited statistically signi?cant
positive returns in the post-grant period. Interestingly, the cumulative post-grant return
for the 25 non-?led grants is almost 2 percent higher than the late-?led grants. This
result is consistent with the suggestion that those ?rms with the most hidden returns to
executives via backdating have the greatest incentive not to lodge the required notice
Figure 2.
Cumulative average
abnormal stock returns
surrounding executive
options grants partitioned
according to ?ling lag
4.00%
6.00%
7.00%
8.00%
Grants with no filing lag
Grants with a filing lag of 2-days or more
3.00%
5.00%
2.00%
0.00%
–25–23–21–19–17–15–13–11 –9 –7 –5 –3 –1 25 23 21 19 17 15 13 11 9 7 5 3 1
Trading days relative to reported grant date
C
u
m
u
l
a
t
i
v
e
A
v
e
r
a
g
e
A
b
n
o
r
m
a
l
R
e
t
u
r
n
1.00%
–1.00%
–2.00%
Notes: This figure shows the cumulative average abnormal stock returns from 25 days prior to
and 25 days after the date of unscheduled executive option grants made by ASX 200 companies
between January 1, 2005 and June 30, 2007. abnormal returns are calculated using a market
model of normal returns based on a five-month estimation period ending four months prior to
the grant date; grant dates are taken from company end-of-year reports; the option grants have
been partitioned according to their filing lag, which is the number of trading days between the
grant date as reported in the company end-of-year report and the date on which the ASX
notified the market via the company announcements pages of the ASX web site:
www.asx.com. au of an Appendix 3B filed in relation to the grant.
Managerial
opportunism
in Australia
35
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of the option grant. Additionally we also show that when the 25 non-?led grants are
added to the sub-sample of late-?led grants, the statistically and economically
signi?cant returns in the post-grant period are con?rmed[14].
5. Conclusion
Motivated by international evidence of managerial opportunism in the granting of
options to company executives, and consequential breach of disclosure requirements in
terms of the reporting of the characteristics of those options, we collect an initial sample
of 161 unscheduled Australian executive stock options granted by ASX 200 ?rms
between January 1, 2005 and June 30, 2007. Consistent with opportunism, be it
spring-loading, backdating or both, we document a positive average abnormal return in
the post-grant period. Despite the ASXdisclosure requirement that Australian ?rms ?le
their executive options “at the same time” as they grant them, we observe that a
signi?cant number of the sampled options were ?led some days after they were granted.
Given that backdated options always have a lag between their grant dates and ?ling
dates, this evidence of late-?ling opens up the possibility that options in the sample
might have been backdated.
Testing directly for evidence of backdating, we observe a positive post-grant average
abnormal return for a sub-sample of late-?led options but no post-grant average
abnormal return for a sub-sample of options ?led on time. In that the grants in both
sub-samples could have been spring-loaded, yet only the late-?led grants could have
been backdated, this ?nding is consistent with the sampled ?rms backdating rather
Same-day ?ling Filing lag $2 days
n ¼ 25 n ¼ 27
Window Mean p-value Mean p-value
25 20.0089 0.0325
* *
20.0026 0.6449
24 20.0019 0.6244 0.0069 0.1747
23 0.0113 0.0310
* *
0.0013 0.7540
22 0.0043 0.3920 0.0049 0.1804
21 20.0057 0.2047 0.0035 0.4200
0 0.0031 0.5474 0.0076 0.1244
1 20.0017 0.6870 0.0052 0.2952
2 0.0040 0.3583 0.0050 0.5391
3 20.0051 0.2426 0.0083 0.1521
4 0.0092 0.0771
*
20.0006 0.8821
5 20.0030 0.4759 0.0079 0.0698
*
(225,0) 0.0099 0.6900 0.0261 0.1235
(1,25) 0.0012 0.9610 0.0473 0.0338
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different fromzero; this table presents daily average abnormal returns and cumulative
average abnormal returns surrounding unscheduled executive option grants made by ASX 200
companies between January 1, 2005 and June 30, 2007; abnormal returns are estimated using a market
model based on a ?ve-month estimation period ending four months prior to the grant date; grant dates
are taken from company end-of-year reports; the option grants have been partitioned according to their
?ling lag, which is the number of trading days between the grant date as reported in the company end-of-
year report and the date on which the ASXnoti?ed the market via the company announcements pages of
the ASX web site: www.asx.com.au of an Appendix 3B ?led in relation to the grant
Table IV.
Abnormal stock returns
surrounding executive
options grants
partitioned according to
?ling lag
ARJ
25,1
36
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
than spring-loading. This result is even stronger when we extend our analysis to those
option grants made by ?rms with a history of late-?ling but for which no notice was ?led
with the ASX. In that ?rms breach the ASX requirement of simultaneous ?ling in order
to facilitate backdating, we suggest that a more stringent enforcement of the continuous
disclosure rules are warranted.
Of additional interest to regulators around the world, this is the ?rst reported case of
backdating in a market where participants were subject to IFRS reporting requirements.
Speci?cally, our study shows that when executive options carry a level of compensation
expense equal to their grant date fair value, ?rms are still motivated to backdate options,
so as to understate the generosity of the executives’ option grants and/or to reduce the
level of expense recorded in their income statement.
Notes
1. See for example the list of ?rms withoptions-grantingpractices that have attractedthe scrutiny
of regulators as reportedbythe Wall Street Journal intheir “Perfect payday: options Scorecard”,
available at: www.online.wsj.com/public/resources/documents/info-optionsscore06-full.html
(accessed July 29, 2011).
Non-?led grants made by ?rms
identi?ed as being late-?lers
Late-?led grants and non-?led
grants made by ?rms identi?ed
as being late-?lers
n ¼ 25 n ¼ 52
Window Mean p-value Mean p-value
25 0.0000 0.9946 20.0014 0.6706
24 20.0031 0.3262 0.0021 0.4898
23 0.0029 0.4871 0.0021 0.4784
22 0.0105 0.0203
* *
0.0076 0.0081
21 0.0056 0.2156 0.0045 0.1445
0 0.0048 0.4063 0.0063 0.0943
*
1 0.0037 0.3076 0.0045 0.1451
2 0.0010 0.8611 0.0031 0.5363
3 20.0033 0.3586 0.0028 0.4287
4 20.0045 0.1888 20.0025 0.3463
5 0.0016 0.7369 0.0048 0.1278
(225,0) 0.0467 0.0299
* *
0.0360 0.0074
* *
(1,25) 0.0651 0.0121
* *
0.0558 0.0009
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different from zero; this table presents daily average abnormal returns and
cumulative average abnormal returns surrounding unscheduled executive option grants made by ASX
200 companies between January 1, 2005 and June 30, 2007; abnormal returns are estimated using a
market model based on a ?ve-month estimation period ending four months prior to the grant date;
grant dates are taken from company end-of-year reports; the option grants have been partitioned into
two sets ?rst is a sub-sample of options for which no Appendix 3B was ?led in relation to the grant on
the company announcements pages of the ASX web site: www.asx.com.au granted by ?rms which had
also granted at least one option deemed to have been ?led with a lag of two or more days (where the
?ling lag is the number of trading days between the grant date as reported in the company end-of-year
report and the date on which the ASX noti?ed the market via the company announcements pages of
the ASX web site (www.asx.com.au), and second is a merged sample of ?rst and of those option grants
which had been recorded as having a ?ling-lag of two or more days
Table V.
Abnormal stock returns
surrounding executive
options grants inclusive
of grants for which no
?ling was recorded
Managerial
opportunism
in Australia
37
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. The fact that Lie (2005) still observed the v-shaped pattern of abnormal returns for the sample
of scheduled option grants suggests that the scheduled sample picked up some grants which
appear to have been backdated. This is not a surprising result given the criterion used
for classifying scheduled grants. Certainly ?rms which grant executive options in an
unscheduled fashion might backdate an option such that its false grant date, by chance, fell
within one week of the ?rst anniversary of a previous option granted by that same ?rm. Even
?rms that did grant at ?xed annual meetings also could have backdated over small enough
windows such that their grants still appeared to be scheduled according to the criterion.
Lie (2005) showed that when the criterion or classifying scheduled grants was tightened
to one day rather than one week around the anniversary of a previous grant, the abnormal
return pattern for the sample of scheduled options disappeared. Accordingly the tighter
criterion appeared more effective in not omitting potentially backdated grants from an
unscheduled sample.
3. In response to the ?ndings in Lie (2005) and Heron and Lie (2007) utilized the tighter criterion
for classifying scheduled grants: grants were deemed scheduled if they fell within one day of
the anniversary of a previous grant made by the same ?rm.
4. ASX Listing Rule 3.10.5 stipulates that listed Australian ?rms are required to notify the ASX
of options grants “at the same time” as they are granted by ?ling an Appendix 3B. By
contrast, since the implementation of the SOA (effective August 29, 2002), US ?rms must ?le
option grants with the SEC “before the end of the second business day” following their
granting.
5. AASB 2: Share-based Payment is equivalent to the international accounting standard IFRS
2: Share-based Payment. AASB 2 is effective for reporting periods after January 1, 2005 and
its application is mandatory for executive options granted after November 7, 2002 that also
vest subsequent to January 1, 2005. The grant date fair value of options is typically adjusted
where the option’s vesting is subject to any performance hurdles. The adjustments will
re?ect the estimated probability that the hurdles will be met. The adjusted aggregate fair
value is then expensed evenly over the option’s vesting period. Since June 15, 2005, US ?rms
have also been required to expense the grant date fair value of their options in accordance
with FASB statement 123 Accounting for Stock-Based Compensation.
6. Aboody and Kasznik (2000) did however document that US ?rms made opportunistic
disclosures of material information around the grant dates of scheduled options. In spite
of this we do not retain scheduled options to investigate this manner of opportunism.
Our chief interest is in the possibility of either backdating or spring-loading affecting
unscheduled grants.
7. This criterion for classifying scheduled options is the same as that employed by Heron
and Lie (2007), which they argued was superior to the less restrictive criterion utilized in
Lie (2005).
8. For each test that we ultimately conduct, the model for normal returns was twice
re-speci?ed, ?rst based upon market-adjusted returns (assuming a zero-one market model)
and second using simple raw returns. In every case the results produced under both of these
alternate models are qualitatively similar to the results produced under the market model.
9. Heron and Lie (2007) observed negative abnormal pre-grant returns for a sample of
663 post-SOX grants with an illegal ?ling lag of four days or more. The majority of grants in
this sample were ?led with the SEC in excess of ten days after they were granted.
10. Uylangco et al. (2010) similarly report relatively high levels of apparent non-compliance with
ASX disclosure requirements in their analysis of directors trading in the shares of their
respective ?rms. If indeed no ?lings were submitted for the options for which ?lings are
missing, then the ASX’s ?rst exposure to these options may well have come from the same
ARJ
25,1
38
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
end-of-?nancial-year reports from which we sampled the options. This opens up the
possibility that these options might have been backdated over extremely large windows. In
results reported in Section 4 of the paper, we show that the sub-sample of non-?led options
which were granted by ?rms also shown to be late-?lers exhibits abnormal returns
consistent with backdating. By contrast the sub-sample of the remaining non-?led options
does not. Accordingly it would appear that backdating could be a factor in explaining why
some of the non-?led options were not ?led. Whilst many ?rms choose to report a grant date
for the options granted in the Appendix 3Bs, it is not actually a requirement for ?rms to
do so. It is for this reason that the grant dates for the options in our sample were taken from
annual end-of-?nancial-year reports (where an option’s grant date must be quoted as an input
used in calculating the compensation expense incurred in granting the option). As required
by ASX Listing Rule 3.10.5, grant ?lings must be made at the same time as the grants
themselves. The fact however that the grant date is not a required ?eld in the Appendix 3B
means that a ?rm’s compliance with ASX Listing Rule 3.10.5 can only be determined once
the end-of-?nancial-year reports are available.
11. Firms that ?led Appendix 3Bs in relation to sampled executive option grants naturally dated
these ?lings. The date on which the ASX made the announcement of having received a given
?ling almost always matched the date on the ?ling itself. However, in a very small number of
cases the announcement came one day later. It is presumed that these discrepancies arose
from ?rms ?ling Appendix 3Bs at the close of trade and the ASX not being able to announce
until the following day that they had received those ?lings. A less likely explanation would
involve ?rms seeking to disguise their own backdating by listing the previous days date on
their Appendix 3Bs when submitting them. Due to this possibility, however remote, we
elected to measure the lag between an option’s grant date and its ?ling date according to
the number of trading days between the grant date (as reported in the annual report) and the
date on which the ASX made the announcement that the option had been ?led, rather than
the date which appeared on the ?ling itself. While this choice may have resulted in a very
small number of grants being recorded as having been ?led one day later than they actually
were, this would, if anything, be expected to bias against ?nding evidence of backdating in a
sample of late-?led grants.
12. As stated previously, the choice to measure the ?ling lag of an option according to the
number of trading days between its grant date (as reported in the annual reports) and
the date on which the ASX made the announcement that the option had been ?led, may have
resulted in a small number of grants being recorded as having been ?led one day later than
they actually were. In light of this possibility, the sub-sample of grants observed to have a
one day ?ling lag may well include grants that were actually ?led on time. Accordingly we
see this sub-sample as un?t to use in our direct test of backdating. Unreported analysis
for the sub-sample revealed that there was no evidence of any systematic abnormal returns
over the event windows examined.
13. While there are individual days in the reported event window where returns were signi?cant
for the same-day ?ling sub-sample, it is notable that these returns are offsetting and that in
aggregate the returns over the pre and post-grant windows are not signi?cantly different to
zero. In contrast, the process of aggregation has strengthened the signal to noise ratio for the
late-?ling sub-sample where there is clear evidence of statistically signi?cant returns in the
post-grant window.
14. When we repeat this analysis for the sub-sample of the remaining 63 non-?led options
granted by ?rms without a record of late-?ling, we ?nd no evidence of statistically
signi?cant positive returns in the post-grant period. These results are available from the
authors upon request.
Managerial
opportunism
in Australia
39
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
References
Aboody, D. and Kasznik, R. (2000), “CEO stock option awards and the timing of corporate
voluntary disclosures”, Journal of Accounting and Economics, Vol. 29 No. 1, pp. 73-100.
Bebchuk, L., Grinstein, Y. and Peyer, U. (2010), “Lucky CEOs and lucky directors”, Journal of
Finance, Vol. 65 No. 6, pp. 2363-401.
Bizjak, J., Lemmon, M. and Whitby, R. (2009), “Option backdating and board interlocks”,
Review of Financial Studies, Vol. 22 No. 11, pp. 4821-47.
Collins, D., Gong, G. and Li, H. (2009), “Corporate governance and backdating of executive stock
options”, Contemporary Accounting Research, Vol. 26 No. 2, pp. 403-45.
Forell, C. and Bandler, J. (2006), “Perfect payday”, Wall Street Journal, March 18, p. 1.
Heron, R. and Lie, E. (2007), “Does backdating explain the stock price pattern around executive
stock option grants?”, Journal of Financial Economics, Vol. 83 No. 2, pp. 271-95.
Lie, E. (2005), “On the timing of CEO stock option awards”, Management Science, Vol. 51 No. 5,
pp. 802-12.
Uylangco, K., Easton, S. and Faff, R. (2010), “The equity and ef?ciency of the Australian share
market with respect to director trading”, Accounting Research Journal, Vol. 23 No. 1,
pp. 5-19.
Yermack, D. (1997), “Good timing: CEO stock option awards and company news
announcements”, Journal of Finance, Vol. 52 No. 2, pp. 449-76.
Further reading
Bernile, G. and Jarrell, G. (2009), “The impact of the options backdating scandal on shareholders”,
Journal of Accounting and Economics, Vol. 47 No. 1, pp. 2-26.
About the authors
Andrew Trumble is a Research Assistant for the Department of Finance at The University of
Melbourne. Additionally he has acted as a Student Tutor at the University for the ?nal year
undergraduate unit Corporate Finance. In 2008, he completed an honours thesis in Finance.
Sean Pinder lectured at both the undergraduate and postgraduate levels in the Department of
Accounting and Finance at the University of Newcastle prior to taking up this position in 2001,
and prior to that at Monash University. He has also taught at the postgraduate level at Lancaster
University in England, the Melbourne Business School and the Adelaide Graduate School of
Business. In 2007 he was awarded the Pearson Education Accounting/Finance Lecturer of the
Year Award. He is co-author of the widely used McGraw-Hill textbook, Business Finance. Now in
its 11 edition, the text has in the past been awarded an Australian Award for Excellence in
Educational Publishing by the Australian Publishers Association. In 2009 he was awarded a
Citation for Outstanding Contributions to Student Learning by the Australian Learning and
Teaching Council for the development and implementation of numerous effective innovations in
teaching ?nance and the successful engagement with practitioners in these innovations. He has
an extensive research pro?le with his work appearing in leading Australian and international
journals and has received a number of prizes for his research. Sean Pinder is the corresponding
author and can be contacted at: [email protected]
ARJ
25,1
40
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Richard Kent, James Routledge. 2015. Use of benchmarks in predicting earnings management?. Accounting
& Finance n/a-n/a. [CrossRef]
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_475785166.pdf
The purpose of this paper is to test for managerial opportunism, specifically the
backdating of executive options, in Australia.
Accounting Research Journal
Evidence of managerial opportunism in Australia
Andrew Trumble Sean Pinder
Article information:
To cite this document:
Andrew Trumble Sean Pinder, (2012),"Evidence of managerial opportunism in Australia", Accounting
Research J ournal, Vol. 25 Iss 1 pp. 25 - 40
Permanent link to this document:http://dx.doi.org/10.1108/10309611211244492
Downloaded on: 24 January 2016, At: 21:14 (PT)
References: this document contains references to 10 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 2635 times since 2012*
Users who downloaded this article also downloaded:
Kim Ittonen, (2012),"Market reactions to qualified audit reports: research approaches", Accounting
Research J ournal, Vol. 25 Iss 1 pp. 8-24http://dx.doi.org/10.1108/10309611211244483
Irene Tempone, Marie Kavanagh, Naomi Segal, Phil Hancock, Bryan Howieson, J enny Kent,
(2012),"Desirable generic attributes for accounting graduates into the twenty-first century:
The views of employers", Accounting Research J ournal, Vol. 25 Iss 1 pp. 41-55 http://
dx.doi.org/10.1108/10309611211244519
Ismail Cebeci, (2012),"Integrating the social maslaha into Islamic finance", Accounting Research J ournal,
Vol. 25 Iss 3 pp. 166-184http://dx.doi.org/10.1108/10309611211290158
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Evidence of managerial
opportunism in Australia
Andrew Trumble and Sean Pinder
Department of Finance, University of Melbourne,
Parkville, Australia
Abstract
Purpose – The purpose of this paper is to test for managerial opportunism, speci?cally the
backdating of executive options, in Australia.
Design/methodology/approach – The paper analyses the return behaviour associated with a
sample of 161 unscheduled options granted by Australian ?rms. Speci?cally, the authors test for
differences between a subsample of grants that had late-?led notices (and hence may be subject to
backdating) versus those that had notices ?led on-time.
Findings – Consistent with backdating, it is found that these abnormal post-grant returns persist for
a sub-sample of late-?led grants but not for a sub-sample of grants with same-day ?ling. Furthermore
– the authors ?nd even stronger results for option grants made by ?rms with a history of late-?ling
but for which no notice was ?led with the Australian Securities Exchange. This paper is the ?rst to
demonstrate these effects in a setting subject to the IFRS requirement that the fair value (rather than
the intrinsic value) of executive options be expensed.
Originality/value – This paper is the ?rst to demonstrate these effects in Australia and further in a
setting subject to the IFRS requirement that the fair value (rather than the intrinsic value) of executive
options be expensed.
Keywords Options backdating, Managerial opportunism, Executive compensation, Australia, Pay
Paper type Research paper
1. Introduction
As a form of performance-sensitive remuneration, executive options have become a
popular means of aligning the interests of rent-extracting executives and pro?t-seeking
shareholders. A stream of US-based literature, however, details how executive option
grants have been used to covertly endow their recipients with performance-insensitive
pay. Yermack (1997) documents positive average abnormal returns over the period of
time immediately following the grant dates of CEO options. He interprets this as
evidence that executives “spring-load” their option grants by opportunistically timing
them to precede foreseeable stock price appreciations. Lie (2005) observes a negative
average abnormal return over the period of time preceding the grant dates of CEO
options and a positive average abnormal return immediately following these dates.
He contends that these abnormal returns are likely attributable to undisclosed “options
backdating”.
Options backdating refers to the practice by which a ?rm falsely claims to have
granted executive options on a past date. The past date nominated will be a date on
which the ?rm’s stock price was lower than the date on which the ?rm makes the false
claim (the current date). This way, secretly backdated options, which appear to have
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
The authors would like to thank Rob Brown, John Handley, Steve Easton and David Yermack for
their helpful comments and advice on previous versions of this paper.
Managerial
opportunism
in Australia
25
Accounting Research Journal
Vol. 25 No. 1, 2012
pp. 25-40
qEmerald Group Publishing Limited
1030-9616
DOI 10.1108/10309611211244492
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
been endowed with a given amount of intrinsic value on the past date, assume
greater intrinsic value as of the current date. Undisclosed backdating of this nature
thereby allows ?rms to understate the true intrinsic value of their executives’ option
grants.
In 2006, backdating became front-page news following a report in The Wall Street
Journal that the practice was endemic within corporate America (Forell and Bandler,
2006). The Securities and Exchange Commission (SEC) and the Department of Justice
have since come to view the practice as fraudulent and in a number of well publicized
cases both civil and criminal charges have been laid[1]. Despite the pro?le of the
backdating problem in the USA, there is no published evidence that the practice also
exists in Australia. Under the Australian Securities Exchange (ASX) Listing Rules
relating to continuous disclosure, ?rms must ?le notice of option grants “at the same
time” as they are granted (ASX Listing Rule 3.10.5), which should make backdating
impracticable. That is, the options will always be reported as having been granted on a
past date when ?ling a grant of backdated options. As such, all backdated options would
have the appearance of being ?led some time after they were reportedly granted and
thus would appear to breach Listing Rule 3.10.5. Notably however, Heron and Lie (2007)
demonstrated that following the implementation of the Sarbanes Oxley Act (SOX) in
2002, US ?rms consistently de?ed the SEC requirement of ?ling notice of options grants
within two days of their granting. The present study therefore investigates whether
Australian ?rms have a similar record of late-?ling which, if demonstrated, implies that
backdating in Australia is a distinct possibility.
We gather a sample of 161 unscheduled stock options granted to the key executives
(including but not limited to the CEO) of 61 ASX200 companies between January 1, 2005
and July 30, 2007. Consistent with some form of opportunism, be it spring-loading or
backdating, we observe a positive average abnormal return of 2.02 percent in the 25-day
post-grant window. A test to determine if this abnormal return pattern is attributable
speci?cally to backdating is made worthwhile given that a number of the sampled
grants were discovered to have been late-?led. Categorising the sampled options
according to those which had same-day ?lings and those which were ?led a minimumof
two days late, we observe that a positive post-grant average abnormal return is retained
for the sub-sample of late-?led options, but disappears for the sub-sample of options ?led
on time. Where the grants in both of these sub-samples could have been spring-loaded,
but only the late-?led grants could have been backdated, these results are consistent
with backdating rather than spring-loading being the key form of opportunism
prevalent in the sample.
The remainder of the paper proceeds as follows. Section 2 provides an overview of
the extant literature which has focused on identifying irregularities in stock price
patterns surrounding the grant dates of executive options as well as an analysis of the
mechanics of options backdating, and details the motivations for backdating as
suggested by US research. This section also documents the differences between the US
and Australian disclosure requirements and accounting standards which pertain to the
practice of backdating. Section 3 details the test for evidence of general opportunism in
the sample of executive options grants, while Section 4 presents the direct test for
evidence of options backdating and then provides a test of the robustness of any
results relating to an increased sample of option grants. Section 5 outlines the
conclusions of the study.
ARJ
25,1
26
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. Options backdating
2.1 Prior literature
The literature concerning stock return patterns surrounding executive option grants
remains almost exclusively US-based. Yermack (1997) provided the ?rst evidence that
option grants were providing a means for rent extraction. Sampling Fortune
500 companies from 1992 to 1994, Yermack found that ?rms making stock option
awards to their CEOs outperformed the market on a risk-adjusted basis by slightly
more than 2 percent during the 50-day post-grant period. Yermack saw this result as
consistent with his hypothesis that CEOs were setting the terms of their own
remuneration packages. Speci?cally, Yermack suggested that executives were having
their option grants timed to be awarded just prior to announcements of favourable
company news, a practice which has become known as spring-loading. Yermack
further substantiated his claim that the observed abnormal return pattern re?ected
undue managerial in?uence by demonstrating that where a granting ?rm’s CEO sat on
its remuneration committee, the 50-day post-grant average abnormal return rose to
11.2 percent.
Aboody and Kasznik (2000) noted the way in which many ?rms scheduled
compensation committee meetings – at which executive options were known to be
granted – to approximately the same date each year. Where these meetings were
scheduled up to a year in advance, the authors noted that options granted at the
meetings could not have been spring-loaded. Accordingly, in analyzing return patterns
surrounding CEO option grants made between 1992 and 1996, Aboody and Kasznik
categorised options as either “scheduled” or “unscheduled” to distinguish between
options which appeared to have been granted on a ?xed annual basis at scheduled
compensation committee meetings, and options which did not. Speci?cally, the study
classi?ed scheduled options as those which had grant dates within one week of the one
year anniversary of a previous grant made by the same ?rm. For the unscheduled
sub-sample, where spring-loading was still thought possible, the study not only found
a positive mean abnormal return in the 30-day post-grant window, but also a negative
mean abnormal return in the 30-day pre-grant window. The study also revealed a
positive mean abnormal return in the 30-day post-grant window for the sub-sample of
scheduled grants. This result led the authors to conclude that while CEOs may not
have been able to opportunistically time the grant dates of scheduled options, they
were still able to in?uence the payoff on these options by leaking price sensitive
information on the days surrounding their grant dates.
Lie (2005) was the ?rst study to empirically identify the practice of options
backdating. Partitioning option grants made between 1992 and 2002 into scheduled
and unscheduled samples (using the same criteria as Aboody and Kasznik (2000)),
Lie found that while both samples exhibited a v-shaped pattern of negative pre-grant
and positive post-grant average abnormal returns, the pattern for the sample of
unscheduled grants was far more pronounced. Lie noted that the price movements
surrounding the dates of the unscheduled option grants were being driven by
market-wide factors which managers were unlikely to have been predicting. This
meant that it was unlikely that spring-loading was behind the abnormal pattern of
returns. As a means of explaining the return patterns, Lie instead suggested that in
order to secretly endow executive options with hidden amounts of intrinsic value, ?rms
may have been retroactively declaring option grants to have been made on past dates
Managerial
opportunism
in Australia
27
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
on which their company stock price was low. In that this “backdating” could not
have been effected on options with grant dates that were thought to be ?xed in
advance, this theory provided an elegant explanation as to why the study found
a stronger abnormal return pattern for the unscheduled sample than it did for the
scheduled sample[2].
Heron and Lie (2007) conducted further tests for evidence of backdating. Following
the implementation of Sarbanes Oxley, effective as of August 29, 2002, the acceptable
post-grant delay for ?ling – and thus the backdating window for any ?rm that wanted
to ?le on time – was reduced signi?cantly to just two business days. Heron and
Lie hypothesized that if backdating had caused the v-shaped pattern of abnormal
returns observed for the sample of unscheduled pre-SOX grants in Lie (2005), then
any abnormal return pattern observable for a sample of post-SOX grants would
likely be much weaker. To test their hypothesis, Heron and Lie gathered two samples
of unscheduled options, a post-SOX sample spanning from August 29, 2002 to
November 30, 2004 and a sub-sample of the pre-SOX grants sampled in Lie (2005)[3].
They found that “the magnitude of the average abnormal return during the week
before (after) the grants (was) roughly six (?ve) times larger” for the pre-SOX sample
than it was for the post-SOX sample (Heron and Lie, 2007, p. 273).
Heron and Lie then sought to con?rm that this weaker but still signi?cant post-SOX
return pattern was attributable to backdating rather than perhaps spring-loading. The
post-SOXoptions were grouped according to the number of days lag between their grant
dates and ?ling dates. Heron and Lie conjectured that if backdating had continued
post-SOX, the abnormal return patterns for the sub-samples would become more
pronounced as the lag increased and the backdating windowwidened. The study found
that a sub-sample of options with a one day ?ling lag exhibited no average abnormal
returns on the days surrounding the grants, but that a sub-sample of grants with a
two-daylagexhibiteda positive dayone average abnormal return. Further, a sub-sample
with a three-day lag exhibited positive average abnormal returns on both day one and
day two, while a sub-sample of grants with a lag of four or more days exhibited a total
re-emergence of the v-shaped pattern of average abnormal returns in all of the event
windows examined. These results not only suggested that US ?rms, post-SOX, deemed
it worthwhile to backdate over windows as small as two days, but also that they were
prepared to breach the two-day ?ling requirement in order to backdate over longer
periods.
2.2 The mechanics of, and motivation for, options backdating
Options backdating refers to the practice by which a ?rm falsely claims to have
granted executive options on a past date. The past date nominated will be a date on
which the ?rm’s stock price was lower than the date on which the ?rm makes the false
claim (the current date). In this situation, secretly backdated options said to have been
granted at-the-money (for example) on the nominated past date will accordingly be
in-the-money as at the current date. Where a ?rm keeps such backdating secret, it will
appear to outsiders (shareholders and regulators included) that the ?rm granted
options with a lower intrinsic value than their actual worth.
Listed ?rms in both the USAand Australia are required to submit ?lings of executive
option grants to their securities exchanges within a set timeframe following the grant
date. Presently, US ?rms have two business days, while Australian ?rms are expected
ARJ
25,1
28
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
theoretically to ?le on the day of the grant[4]. Because backdating involves nominating a
past date as the date on which an option was said to have been granted, there will always
be a lag between the nominated grant date of a backdated option and the date on which
the option is ultimately ?led. Therefore, if a ?rm wishes to backdate an option but also
wants to be seen to be ?lingontime, the windowover whichthe ?rmcanbackdate is only
as long as the permissible lag in ?ling. For example, consider the US ?rm has a two-day
window in which to ?le options following their granting. If the ?rm secretly backdated
an option by two days and then ?led immediately, the exchange would be left to believe
that the ?rm had physically granted the option two days earlier and had simply waited
its full measure before ultimately ?ling on time. By contrast, as Australian ?rms are
required to ?le notice of options immediately upon their granting, it would be impossible
to both backdate options and meet the ?ling requirement. Australian backdaters
therefore could not avoid appearing as late-?lers.
Prior to June 15, 2005, the accounting treatment for executive stock option grants in
the USA was governed by Accounting Principles Board (APB) Opinion Number
25 Accounting for Stock Issued to Employees. Opinion 25 required US ?rms to expense
the grant date intrinsic value of any executive options granted; options secretly
backdated to appear at-the-money therefore carried no compensation charge for the
?rm. Where these backdated at-the-money options would have intrinsic value as of
their real grant dates, substituting them for other forms of executive compensation
would have provided a means for US ?rms to reduce their remuneration expense and
boost reported pro?t without necessarily reducing the effective level of payout to their
executives. Had the US ?rms that backdated done so for this purpose, it would have
been expected that, relative to their peers, they would have paid lower levels of other
forms of executive compensation. However, in what is a telling result, Bebchuk et al.
(2010) found that relative to their peers, US ?rms suspected of backdating over the
period 1996-2005 were in fact paying higher levels of executive compensation in other
forms.
Given that backdated options therefore appear to be adding to, rather than
substituting for, other forms of executive remuneration, undisclosed backdating has
come to be seen as a means of rent extraction symptomatic of executives exerting
excessive in?uence over their own remuneration. Consistent with this managerial
power theory, a number of papers (Bizjak et al., 2009; Collins et al., 2009; Bebchuk et al.,
2010) demonstrate that a variety of ?rm characteristics conducive to excessive CEO
power and indicative of poor governance were signi?cant in determining which ?rms
appeared to be backdating their grants. For example, Collins et al. (2009) showed that
the likelihood of backdating increased with the percentage of inside directors on the
board, the presence of inside directors on the audit committee, the presence of outside
directors hired by the incumbent CEO, and where the CEO doubled as chairperson.
In what is a particularly striking result, the study also showed that a CEO’s grant was
more likely to have been backdated where an additional grant was made to a company
director on the same-day.
2.3 Backdating in Australia
Our sample consists of executive options granted in Australia after January 1,
2005. There are important differences between this Australian environment and
the pre-2005 US environment that has been the focus of extant studies into
Managerial
opportunism
in Australia
29
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
options backdating. We demonstrate that not only do these differences not necessarily
preclude the possibility of backdating in our sample but also that our sampled ?rms
and their executives remain in a position to realise large ?nancial bene?ts as a result
of options backdating.
2.3.1 Australian ?ling requirements. Under the ASX continuous disclosure
requirements, ASX Listing Rule 3.10.5 stipulates that listed Australian ?rms, through
the ?ling of an “Appendix 3B: new issue announcement: application for quotation of
additional securities and agreement” (Appendix 3B) notice, are required to notify the
ASX of option grants “at the same time” as they are granted. Notably, because
backdaters reporting their grants choose to cite a past date as the grant date, any
backdated options in Australia would appear to be late-?led and in breach of Listing
Rule 3.10.5. Our investigation into Australian backdating would therefore have been
made redundant if ?rms strictly adhered to the requirement. Importantly, though,
it appears they do not. Speci?cally, as we report in Section 3 of the paper, we observe
that an Appendix 3B was ?led for only approximately 44 percent of the sampled option
grants and that only about a third of these Appendix 3Bs were ?led on the same date
that the options were reported to have been granted.
2.3.2 Australian accounting standards. The expensing requirement for employee
options determines howconcealing an option’s intrinsic value via backdating might also
result in ?rms under-reporting the appropriate level of remuneration expense.
Fundamentally, the requirement will not alter a backdaters ability to conceal intrinsic
value – and to this end will not in?uence the primary incentive to backdate.
The expensing requirement for Australian ?rms in our sample is slightly different from
the requirement imposed upon the US ?rms sampled in prior studies. Speci?cally, prior
to June 15, 2005, APB Opinion Number 25, Accounting for Stock Issued to Employees
required US ?rms to expense the grant date intrinsic value of their executive options.
In contrast, our post January 1, 2005 sample of ?rms were required to expense the grant
date fair value of their executive options, as per AASB 2: Share-based Payment[5].
Despite this slight difference, because backdating involves understating the actual
intrinsic value and thus also the actual fair value of options, both US and Australian
backdaters would have reported lower compensation expense and boosted their
reported pro?tability, while at the same time concealing from the market the true value
of the compensation package provided to their executives.
3. Test for evidence of general opportunism
3.1 Sample
We sample options granted to company executives – including but not limited to the
CEO – of ?rms in the ASX200 as at June 30, 2007. As we only include option grants that
were subject to International Financial Reporting Standards (IFRS), the starting date for
the sample is January 1, 2005.
The sampled options, speci?cally their grant dates and exercise prices, are taken
from company end-of-?nancial-year reports. If a company made multiple grants on the
same date with different exercise prices, we sample only the grant with the lowest
exercise price. If a company made multiple grants on the same date with the same
exercise price (but perhaps with different vesting schedules) we randomly sample only
one of these grants.
ARJ
25,1
30
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
In the USA, the compensation committee meeting at which a ?rm’s executive
options are granted is typically scheduled to fall on approximately the same date each
year. Because the grant dates of options granted at these meetings are ?xed, such
“scheduled” options are not deemed susceptible to spring-loading or backdating[6]. For
the Australian ?rms whose annual reports we survey, a number of their reported
options similarly appear to have been granted on an annual schedule and are thus
excluded from our investigation into opportunism. Speci?cally, we deem an option to
have been scheduled, and thus omit it from the sample, if we observe that another
grant has been made by the same ?rm within one day of the anniversary of the option,
either in the year before, or in the year after the option was granted[7].
Table I displays the key summary statistics for the ?nal sample. The sample
comprises 161 unscheduled executive option grants, made by 61 different companies.
The intrinsic value of the options is determined according to the closing stock price on
the reported grant date.
3.2 Methodology
We measure abnormal returns on a continuously compounding basis over a 51-day
event window surrounding the grant dates of the 161 sampled unscheduled options.
Normal returns on a stock are estimated according to a market model comprising stock
betas which were estimated over a ?ve-month estimation window ending
approximately four months prior to the grant date[8]. From Datastream we obtained
total returns data on the value-weighted S&P ASX 200 Index which was taken as a
proxy for the market portfolio. Individual stock return data and standard proxies for
risk-free rates of return were obtained from the IRESS database.
The test for evidence of general opportunism is a general test in that any positive
abnormal pattern in returns could stem from either spring-loading, backdating or both.
Given that the fundamental aim of both backdating and spring-loading is to capture
post-grant stock price appreciations, a failure to observe positive post-grant abnormal
returns would be inconsistent with both practices. In terms of the pre-grant period,
there is no return pattern that would be inconsistent with either practice. It should
be noted that two prominent US studies of backdating – Lie (2005) and Heron and
Lie (2007) – both documented negative average abnormal returns in the pre-grant
period. These results indicate that in striving for greater gains, US ?rms backdated
options to dates at which their company stock price was at a local minimum. Indeed,
Heron and Lie (2007) demonstrated that post-SOX, US ?rms still appeared to backdate
General
Total number of unscheduled grants 161
Total number of ?rms 61
Intrinsic value of options granted
Options granted were out-of-the-money 62
Options granted were at-the-money 10
Options granted were in-the-money 89
Notes: Descriptive statistics for a sample of unscheduled executive stock option grants made by
ASX200 ?rms (according to the composition at June 30, 2007) between January 1, 2005 and June 30,
2007; the intrinsic value of an option was determined according to the closing stock price on the
reported grant date of the option
Table I.
Descriptive statistics for
sample of unscheduled
executive options grants
Managerial
opportunism
in Australia
31
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
to the local minimum even when this meant backdating by considerably more than two
days, and accordingly breaching the two-day ?ling deadline by considerable margins[9].
This local minimum-at-all-costs approach is not however a necessity when backdating.
Had the US backdaters seen greater risk of detection in ?ling excessively late, then they
may well have elected to stop short of local minimums, such that the pre-grant abnormal
returns might have appeared positive. Ultimately this study will report pre-grant
abnormal returns so as to aid comparability with the prior US studies; however, in
identifying evidence of opportunism, our focus will be on post-grant abnormal returns.
3.3 Results
Figure 1 and Table II display the average abnormal returns over the 51-day window
surrounding the grant dates for the sample of 161 unscheduled option grants.
The positive average abnormal return of 2.02 percent in the 25-day post-grant
period is consistent with the presence of some form(s) of opportunism, be it backdating,
spring-loading or both.
Figure 1.
Cumulative average
abnormal returns
surrounding the grant
date of executive stock
option grants
4.00%
3.50%
3.00%
2.50%
2.00%
1.50%
0.50%
0.00%
–25–23–21–19–17–15–13–11 –9 –7 –5 –3 –1 25 23 21 19 17 15 13 11 9 7 5 3 1
Trading days relative to reported grant date
C
u
m
u
l
a
t
i
v
e
A
v
e
r
a
g
e
A
b
n
o
r
m
a
l
R
e
t
u
r
n
1.00%
All option grants
Notes: This figure shows the cumulative average abnormal stock returns from 25 days prior
to 25 days after the grant date for a sample of unscheduled executive stock option grants made
by ASX 200 companies between January 1, 2005 and June 30, 2007; abnormal returns are
estimated using a market model of normal returns based on a five month estimation period
ending four months prior to the grant date
ARJ
25,1
32
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4. Direct test for evidence of options backdating
4.1 Sample
We are interested in determining whether the average abnormal post-grant return
observed for the sample of executive option grants is attributable to backdating.
Where only late-?led grants can have been backdated, the possibility that backdating
was behind the average abnormal return can only be considered if there is evidence of
late-?ling in the sample. To determine if there is, we search the ASX web site www.asx.
com.au where ?lings of Appendix 3Bs are announced and where copies of the
appendices are available. As shown in Table III, there is a high number of instances
(88 out of 161) where there appears to have been no announcement from the ASX of an
Appendix 3B having been ?led for options granted by a sample ?rm. This result
suggests that either the ASX systematically failed to notify the public of ?lings
received, or perhaps more seriously, that ?rms failed to submit Appendix 3Bs for these
particular options[10]. We therefore measure the ?ling lags for the remaining 73 option
grants for which a relevant Appendix 3B is identi?able. The lag between an option’s
grant date and its ?ling date is calculated according to the number of trading days
between the option’s grant date, as reported in the ?rm’s annual report[11], and the
date on which the ASX announced the option to have been ?led. Table III documents
the ?ling lags and the summary statistics for these 73 option grants.
The ?lings for 27 of these option grants were announced by the ASX two or more
trading days after the options’ respective grant dates. This result clearly suggests that
?rms are contravening the ASX requirement that executive options be ?led “at the
same time” as they are granted and accordingly makes backdating in Australia a
distinct possibility.
All grants
n ¼ 161
Window Mean p-value
25 20.0028 0.1058
24 0.0016 0.3493
23 0.0031 0.0610
*
22 0.0016 0.3120
21 0.0026 0.1212
0 0.0031 0.0992
*
1 0.0027 0.1213
2 0.0015 0.4296
3 0.0012 0.4756
4 0.0003 0.8889
5 20.0001 0.9680
(225,0) 0.0150 0.0883
*
(1,25) 0.0202 0.0305
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different from zero; this table presents daily average abnormal returns and
cumulative average abnormal returns surrounding unscheduled executive option grants made by ASX
200 companies between January 1, 2005 and June 30, 2007; abnormal stock returns are estimated using
a market model of normal returns based on a ?ve month estimation period ending four months prior to
the grant date; grant dates are taken from company end-of-year reports
Table II.
Abnormal stock returns
surrounding executive
options grants
Managerial
opportunism
in Australia
33
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
4.2 Methodology
In testing directly for evidence of backdating, we create two sub-samples, one
comprising grants with same-day ?ling and the other comprising grants ?led two or
more days late[12]. If over the sample period Australian ?rms engaged in only options
backdating, the sub-sample of late-?led options ought to exhibit positive post-grant
abnormal returns, while the sub-sample of options ?led on time, which were not
backdated, should not. Alternatively, if spring-loading occurred then positive
post-grant abnormal returns ought to be observed in both sub-samples given that a
?rm’s choice to spring-load would be independent of its ?ling practices.
4.3 Results
Figure 2 and Table IVdemonstrate that there is a positive post-grant average abnormal
return for the sub-sample of late-?led options but not for the sub-sample of options with
same-day ?ling. Speci?cally, in the ?ve-week period following the grant, ?rms that ?led
grants two or more days late appear on average to have abnormal returns of almost
5 percent, while ?rms that ?led on time realised abnormal returns that, on average, were
not signi?cantly different from zero. This difference is consistent with backdating,
rather than spring-loading, being the key form of opportunism surrounding executive
stock option grants in Australia over the sample period[13].
Panel A: ?ling lags
Filing lag ,0 0 1 2 3 4 5 6 7 8 9 10 11 þ Subtotal
No ?ling
recorded Total
Frequency 2 25 19 4 8 2 1 1 2 1 1 0 7 73 88 161
Panel B: descriptive statistics
General
Total number of
unscheduled
grants 73
Total number of
?rms 33
Intrinsic value of
options granted
Options granted
were out-of-the-
money 29
Options granted
were at-the-
money 5
Options granted
were in-the-
money 39
Notes: Filing lags and descriptive statistics for a sub-sample of unscheduled executive stock option
grants made by ASX200 ?rms (according to the composition at June 30, 2007) between January 1, 2005
and June 30, 2007; the ?ling lag is the number of trading days between the grant date as reported in the
company end-of-year report and the date on which the ASX noti?ed the market via the company
announcements pages of the ASX web site: www.asx.com.au of an Appendix 3B ?led in relation to the
grant; the intrinsic value of an option was determined according to the closing stock price on the
reported grant date of the option
Table III.
Filing lags and
descriptive statistics for
the sample of
unscheduled executive
options grants
ARJ
25,1
34
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Having examined the late-?led options for evidence consistent with backdating,
it is important to also examine the 88 grants for which no Appendix 3B was observed to
have been lodged. As there were no apparent ?lings for these option grants, the ?rst
external noti?cation of them may well have come from the end-of-?nancial-year reports
from which we sampled. If this was the case, then these option grants, like the late-?led
ones, could also have been backdated. While an initial test of the 88-?led options
did not reveal any abnormal returns consistent with backdating, we noted that
25 of the non-?led options had been granted by ?rms that had also granted at least one of
the late-?led option grants. For these 25 options grants, Table V shows
evidence consistent with further options backdating. Speci?cally, the sub-sample of
these 25 grants, like the late-?led sub-sample, also exhibited statistically signi?cant
positive returns in the post-grant period. Interestingly, the cumulative post-grant return
for the 25 non-?led grants is almost 2 percent higher than the late-?led grants. This
result is consistent with the suggestion that those ?rms with the most hidden returns to
executives via backdating have the greatest incentive not to lodge the required notice
Figure 2.
Cumulative average
abnormal stock returns
surrounding executive
options grants partitioned
according to ?ling lag
4.00%
6.00%
7.00%
8.00%
Grants with no filing lag
Grants with a filing lag of 2-days or more
3.00%
5.00%
2.00%
0.00%
–25–23–21–19–17–15–13–11 –9 –7 –5 –3 –1 25 23 21 19 17 15 13 11 9 7 5 3 1
Trading days relative to reported grant date
C
u
m
u
l
a
t
i
v
e
A
v
e
r
a
g
e
A
b
n
o
r
m
a
l
R
e
t
u
r
n
1.00%
–1.00%
–2.00%
Notes: This figure shows the cumulative average abnormal stock returns from 25 days prior to
and 25 days after the date of unscheduled executive option grants made by ASX 200 companies
between January 1, 2005 and June 30, 2007. abnormal returns are calculated using a market
model of normal returns based on a five-month estimation period ending four months prior to
the grant date; grant dates are taken from company end-of-year reports; the option grants have
been partitioned according to their filing lag, which is the number of trading days between the
grant date as reported in the company end-of-year report and the date on which the ASX
notified the market via the company announcements pages of the ASX web site:
www.asx.com. au of an Appendix 3B filed in relation to the grant.
Managerial
opportunism
in Australia
35
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of the option grant. Additionally we also show that when the 25 non-?led grants are
added to the sub-sample of late-?led grants, the statistically and economically
signi?cant returns in the post-grant period are con?rmed[14].
5. Conclusion
Motivated by international evidence of managerial opportunism in the granting of
options to company executives, and consequential breach of disclosure requirements in
terms of the reporting of the characteristics of those options, we collect an initial sample
of 161 unscheduled Australian executive stock options granted by ASX 200 ?rms
between January 1, 2005 and June 30, 2007. Consistent with opportunism, be it
spring-loading, backdating or both, we document a positive average abnormal return in
the post-grant period. Despite the ASXdisclosure requirement that Australian ?rms ?le
their executive options “at the same time” as they grant them, we observe that a
signi?cant number of the sampled options were ?led some days after they were granted.
Given that backdated options always have a lag between their grant dates and ?ling
dates, this evidence of late-?ling opens up the possibility that options in the sample
might have been backdated.
Testing directly for evidence of backdating, we observe a positive post-grant average
abnormal return for a sub-sample of late-?led options but no post-grant average
abnormal return for a sub-sample of options ?led on time. In that the grants in both
sub-samples could have been spring-loaded, yet only the late-?led grants could have
been backdated, this ?nding is consistent with the sampled ?rms backdating rather
Same-day ?ling Filing lag $2 days
n ¼ 25 n ¼ 27
Window Mean p-value Mean p-value
25 20.0089 0.0325
* *
20.0026 0.6449
24 20.0019 0.6244 0.0069 0.1747
23 0.0113 0.0310
* *
0.0013 0.7540
22 0.0043 0.3920 0.0049 0.1804
21 20.0057 0.2047 0.0035 0.4200
0 0.0031 0.5474 0.0076 0.1244
1 20.0017 0.6870 0.0052 0.2952
2 0.0040 0.3583 0.0050 0.5391
3 20.0051 0.2426 0.0083 0.1521
4 0.0092 0.0771
*
20.0006 0.8821
5 20.0030 0.4759 0.0079 0.0698
*
(225,0) 0.0099 0.6900 0.0261 0.1235
(1,25) 0.0012 0.9610 0.0473 0.0338
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different fromzero; this table presents daily average abnormal returns and cumulative
average abnormal returns surrounding unscheduled executive option grants made by ASX 200
companies between January 1, 2005 and June 30, 2007; abnormal returns are estimated using a market
model based on a ?ve-month estimation period ending four months prior to the grant date; grant dates
are taken from company end-of-year reports; the option grants have been partitioned according to their
?ling lag, which is the number of trading days between the grant date as reported in the company end-of-
year report and the date on which the ASXnoti?ed the market via the company announcements pages of
the ASX web site: www.asx.com.au of an Appendix 3B ?led in relation to the grant
Table IV.
Abnormal stock returns
surrounding executive
options grants
partitioned according to
?ling lag
ARJ
25,1
36
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
than spring-loading. This result is even stronger when we extend our analysis to those
option grants made by ?rms with a history of late-?ling but for which no notice was ?led
with the ASX. In that ?rms breach the ASX requirement of simultaneous ?ling in order
to facilitate backdating, we suggest that a more stringent enforcement of the continuous
disclosure rules are warranted.
Of additional interest to regulators around the world, this is the ?rst reported case of
backdating in a market where participants were subject to IFRS reporting requirements.
Speci?cally, our study shows that when executive options carry a level of compensation
expense equal to their grant date fair value, ?rms are still motivated to backdate options,
so as to understate the generosity of the executives’ option grants and/or to reduce the
level of expense recorded in their income statement.
Notes
1. See for example the list of ?rms withoptions-grantingpractices that have attractedthe scrutiny
of regulators as reportedbythe Wall Street Journal intheir “Perfect payday: options Scorecard”,
available at: www.online.wsj.com/public/resources/documents/info-optionsscore06-full.html
(accessed July 29, 2011).
Non-?led grants made by ?rms
identi?ed as being late-?lers
Late-?led grants and non-?led
grants made by ?rms identi?ed
as being late-?lers
n ¼ 25 n ¼ 52
Window Mean p-value Mean p-value
25 0.0000 0.9946 20.0014 0.6706
24 20.0031 0.3262 0.0021 0.4898
23 0.0029 0.4871 0.0021 0.4784
22 0.0105 0.0203
* *
0.0076 0.0081
21 0.0056 0.2156 0.0045 0.1445
0 0.0048 0.4063 0.0063 0.0943
*
1 0.0037 0.3076 0.0045 0.1451
2 0.0010 0.8611 0.0031 0.5363
3 20.0033 0.3586 0.0028 0.4287
4 20.0045 0.1888 20.0025 0.3463
5 0.0016 0.7369 0.0048 0.1278
(225,0) 0.0467 0.0299
* *
0.0360 0.0074
* *
(1,25) 0.0651 0.0121
* *
0.0558 0.0009
* *
Notes: Signi?cant at:
*
10 and
* *
5 percent levels from a two-tailed t-test of whether the mean returns
are signi?cantly different from zero; this table presents daily average abnormal returns and
cumulative average abnormal returns surrounding unscheduled executive option grants made by ASX
200 companies between January 1, 2005 and June 30, 2007; abnormal returns are estimated using a
market model based on a ?ve-month estimation period ending four months prior to the grant date;
grant dates are taken from company end-of-year reports; the option grants have been partitioned into
two sets ?rst is a sub-sample of options for which no Appendix 3B was ?led in relation to the grant on
the company announcements pages of the ASX web site: www.asx.com.au granted by ?rms which had
also granted at least one option deemed to have been ?led with a lag of two or more days (where the
?ling lag is the number of trading days between the grant date as reported in the company end-of-year
report and the date on which the ASX noti?ed the market via the company announcements pages of
the ASX web site (www.asx.com.au), and second is a merged sample of ?rst and of those option grants
which had been recorded as having a ?ling-lag of two or more days
Table V.
Abnormal stock returns
surrounding executive
options grants inclusive
of grants for which no
?ling was recorded
Managerial
opportunism
in Australia
37
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2. The fact that Lie (2005) still observed the v-shaped pattern of abnormal returns for the sample
of scheduled option grants suggests that the scheduled sample picked up some grants which
appear to have been backdated. This is not a surprising result given the criterion used
for classifying scheduled grants. Certainly ?rms which grant executive options in an
unscheduled fashion might backdate an option such that its false grant date, by chance, fell
within one week of the ?rst anniversary of a previous option granted by that same ?rm. Even
?rms that did grant at ?xed annual meetings also could have backdated over small enough
windows such that their grants still appeared to be scheduled according to the criterion.
Lie (2005) showed that when the criterion or classifying scheduled grants was tightened
to one day rather than one week around the anniversary of a previous grant, the abnormal
return pattern for the sample of scheduled options disappeared. Accordingly the tighter
criterion appeared more effective in not omitting potentially backdated grants from an
unscheduled sample.
3. In response to the ?ndings in Lie (2005) and Heron and Lie (2007) utilized the tighter criterion
for classifying scheduled grants: grants were deemed scheduled if they fell within one day of
the anniversary of a previous grant made by the same ?rm.
4. ASX Listing Rule 3.10.5 stipulates that listed Australian ?rms are required to notify the ASX
of options grants “at the same time” as they are granted by ?ling an Appendix 3B. By
contrast, since the implementation of the SOA (effective August 29, 2002), US ?rms must ?le
option grants with the SEC “before the end of the second business day” following their
granting.
5. AASB 2: Share-based Payment is equivalent to the international accounting standard IFRS
2: Share-based Payment. AASB 2 is effective for reporting periods after January 1, 2005 and
its application is mandatory for executive options granted after November 7, 2002 that also
vest subsequent to January 1, 2005. The grant date fair value of options is typically adjusted
where the option’s vesting is subject to any performance hurdles. The adjustments will
re?ect the estimated probability that the hurdles will be met. The adjusted aggregate fair
value is then expensed evenly over the option’s vesting period. Since June 15, 2005, US ?rms
have also been required to expense the grant date fair value of their options in accordance
with FASB statement 123 Accounting for Stock-Based Compensation.
6. Aboody and Kasznik (2000) did however document that US ?rms made opportunistic
disclosures of material information around the grant dates of scheduled options. In spite
of this we do not retain scheduled options to investigate this manner of opportunism.
Our chief interest is in the possibility of either backdating or spring-loading affecting
unscheduled grants.
7. This criterion for classifying scheduled options is the same as that employed by Heron
and Lie (2007), which they argued was superior to the less restrictive criterion utilized in
Lie (2005).
8. For each test that we ultimately conduct, the model for normal returns was twice
re-speci?ed, ?rst based upon market-adjusted returns (assuming a zero-one market model)
and second using simple raw returns. In every case the results produced under both of these
alternate models are qualitatively similar to the results produced under the market model.
9. Heron and Lie (2007) observed negative abnormal pre-grant returns for a sample of
663 post-SOX grants with an illegal ?ling lag of four days or more. The majority of grants in
this sample were ?led with the SEC in excess of ten days after they were granted.
10. Uylangco et al. (2010) similarly report relatively high levels of apparent non-compliance with
ASX disclosure requirements in their analysis of directors trading in the shares of their
respective ?rms. If indeed no ?lings were submitted for the options for which ?lings are
missing, then the ASX’s ?rst exposure to these options may well have come from the same
ARJ
25,1
38
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
end-of-?nancial-year reports from which we sampled the options. This opens up the
possibility that these options might have been backdated over extremely large windows. In
results reported in Section 4 of the paper, we show that the sub-sample of non-?led options
which were granted by ?rms also shown to be late-?lers exhibits abnormal returns
consistent with backdating. By contrast the sub-sample of the remaining non-?led options
does not. Accordingly it would appear that backdating could be a factor in explaining why
some of the non-?led options were not ?led. Whilst many ?rms choose to report a grant date
for the options granted in the Appendix 3Bs, it is not actually a requirement for ?rms to
do so. It is for this reason that the grant dates for the options in our sample were taken from
annual end-of-?nancial-year reports (where an option’s grant date must be quoted as an input
used in calculating the compensation expense incurred in granting the option). As required
by ASX Listing Rule 3.10.5, grant ?lings must be made at the same time as the grants
themselves. The fact however that the grant date is not a required ?eld in the Appendix 3B
means that a ?rm’s compliance with ASX Listing Rule 3.10.5 can only be determined once
the end-of-?nancial-year reports are available.
11. Firms that ?led Appendix 3Bs in relation to sampled executive option grants naturally dated
these ?lings. The date on which the ASX made the announcement of having received a given
?ling almost always matched the date on the ?ling itself. However, in a very small number of
cases the announcement came one day later. It is presumed that these discrepancies arose
from ?rms ?ling Appendix 3Bs at the close of trade and the ASX not being able to announce
until the following day that they had received those ?lings. A less likely explanation would
involve ?rms seeking to disguise their own backdating by listing the previous days date on
their Appendix 3Bs when submitting them. Due to this possibility, however remote, we
elected to measure the lag between an option’s grant date and its ?ling date according to
the number of trading days between the grant date (as reported in the annual report) and the
date on which the ASX made the announcement that the option had been ?led, rather than
the date which appeared on the ?ling itself. While this choice may have resulted in a very
small number of grants being recorded as having been ?led one day later than they actually
were, this would, if anything, be expected to bias against ?nding evidence of backdating in a
sample of late-?led grants.
12. As stated previously, the choice to measure the ?ling lag of an option according to the
number of trading days between its grant date (as reported in the annual reports) and
the date on which the ASX made the announcement that the option had been ?led, may have
resulted in a small number of grants being recorded as having been ?led one day later than
they actually were. In light of this possibility, the sub-sample of grants observed to have a
one day ?ling lag may well include grants that were actually ?led on time. Accordingly we
see this sub-sample as un?t to use in our direct test of backdating. Unreported analysis
for the sub-sample revealed that there was no evidence of any systematic abnormal returns
over the event windows examined.
13. While there are individual days in the reported event window where returns were signi?cant
for the same-day ?ling sub-sample, it is notable that these returns are offsetting and that in
aggregate the returns over the pre and post-grant windows are not signi?cantly different to
zero. In contrast, the process of aggregation has strengthened the signal to noise ratio for the
late-?ling sub-sample where there is clear evidence of statistically signi?cant returns in the
post-grant window.
14. When we repeat this analysis for the sub-sample of the remaining 63 non-?led options
granted by ?rms without a record of late-?ling, we ?nd no evidence of statistically
signi?cant positive returns in the post-grant period. These results are available from the
authors upon request.
Managerial
opportunism
in Australia
39
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
References
Aboody, D. and Kasznik, R. (2000), “CEO stock option awards and the timing of corporate
voluntary disclosures”, Journal of Accounting and Economics, Vol. 29 No. 1, pp. 73-100.
Bebchuk, L., Grinstein, Y. and Peyer, U. (2010), “Lucky CEOs and lucky directors”, Journal of
Finance, Vol. 65 No. 6, pp. 2363-401.
Bizjak, J., Lemmon, M. and Whitby, R. (2009), “Option backdating and board interlocks”,
Review of Financial Studies, Vol. 22 No. 11, pp. 4821-47.
Collins, D., Gong, G. and Li, H. (2009), “Corporate governance and backdating of executive stock
options”, Contemporary Accounting Research, Vol. 26 No. 2, pp. 403-45.
Forell, C. and Bandler, J. (2006), “Perfect payday”, Wall Street Journal, March 18, p. 1.
Heron, R. and Lie, E. (2007), “Does backdating explain the stock price pattern around executive
stock option grants?”, Journal of Financial Economics, Vol. 83 No. 2, pp. 271-95.
Lie, E. (2005), “On the timing of CEO stock option awards”, Management Science, Vol. 51 No. 5,
pp. 802-12.
Uylangco, K., Easton, S. and Faff, R. (2010), “The equity and ef?ciency of the Australian share
market with respect to director trading”, Accounting Research Journal, Vol. 23 No. 1,
pp. 5-19.
Yermack, D. (1997), “Good timing: CEO stock option awards and company news
announcements”, Journal of Finance, Vol. 52 No. 2, pp. 449-76.
Further reading
Bernile, G. and Jarrell, G. (2009), “The impact of the options backdating scandal on shareholders”,
Journal of Accounting and Economics, Vol. 47 No. 1, pp. 2-26.
About the authors
Andrew Trumble is a Research Assistant for the Department of Finance at The University of
Melbourne. Additionally he has acted as a Student Tutor at the University for the ?nal year
undergraduate unit Corporate Finance. In 2008, he completed an honours thesis in Finance.
Sean Pinder lectured at both the undergraduate and postgraduate levels in the Department of
Accounting and Finance at the University of Newcastle prior to taking up this position in 2001,
and prior to that at Monash University. He has also taught at the postgraduate level at Lancaster
University in England, the Melbourne Business School and the Adelaide Graduate School of
Business. In 2007 he was awarded the Pearson Education Accounting/Finance Lecturer of the
Year Award. He is co-author of the widely used McGraw-Hill textbook, Business Finance. Now in
its 11 edition, the text has in the past been awarded an Australian Award for Excellence in
Educational Publishing by the Australian Publishers Association. In 2009 he was awarded a
Citation for Outstanding Contributions to Student Learning by the Australian Learning and
Teaching Council for the development and implementation of numerous effective innovations in
teaching ?nance and the successful engagement with practitioners in these innovations. He has
an extensive research pro?le with his work appearing in leading Australian and international
journals and has received a number of prizes for his research. Sean Pinder is the corresponding
author and can be contacted at: [email protected]
ARJ
25,1
40
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
This article has been cited by:
1. Richard Kent, James Routledge. 2015. Use of benchmarks in predicting earnings management?. Accounting
& Finance n/a-n/a. [CrossRef]
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
4
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_475785166.pdf