Description
This paper aims to examine the relationship between firms’ decisions to expense employee
stock options (ESOs) under the voluntary period of Statement of Financial Accounting Standard No. 123
(SFAS 123) and their market-to-book (MTB-1) ratio and conditional conservatism. Conservatism is
chosen because, even though expensing of ESOs is a conservative accounting treatment, it is not
obvious how the decision would be related to the MTB-1 ratio and conditional conservatism,
particularly because the MTB-1 is a long-run measure of conservatism and the decision to voluntarily
expense is examined over two years. The setting provides a unique opportunity to assess how two
major proxies of conservatism are related.
Accounting Research Journal
Market-to-book ratio and conditional conservatism: firms’ voluntary expensing of
employee stock options
Gulraze Wakil
Article information:
To cite this document:
Gulraze Wakil , (2014),"Market-to-book ratio and conditional conservatism: firms’ voluntary expensing of
employee stock options", Accounting Research J ournal, Vol. 27 Iss 2 pp. 124 - 149
Permanent link to this document:
http://dx.doi.org/10.1108/ARJ -11-2012-0088
Downloaded on: 24 January 2016, At: 21:20 (PT)
References: this document contains references to 48 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 520 times since 2014*
Users who downloaded this article also downloaded:
Prodosh Simlai, (2014),"Firm characteristics, distress risk and average stock returns", Accounting Research
J ournal, Vol. 27 Iss 2 pp. 101-123 http://dx.doi.org/10.1108/ARJ -06-2012-0046
Etumudon Ndidi Asien, (2014),"Impact of firm-specific characteristics on managers’ identity disclosure",
Accounting Research J ournal, Vol. 27 Iss 2 pp. 150-168 http://dx.doi.org/10.1108/ARJ -03-2013-0010
Ning Du, J ohn E. McEnroe, Kevin Stevens, (2014),"The joint effects of management incentive and
information precision on perceived reliability in fair value estimates", Accounting Research J ournal, Vol. 27
Iss 2 pp. 188-206 http://dx.doi.org/10.1108/ARJ -10-2012-0081
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Market-to-book ratio and
conditional conservatism: frms’
voluntary expensing of employee
stock options
Gulraze Wakil
Sprott School of Business, Carleton University, Ottawa, Canada
Abstract
Purpose – This paper aims to examine the relationship between frms’ decisions to expense employee
stock options (ESOs) under the voluntary period of Statement of Financial Accounting Standard No. 123
(SFAS 123) and their market-to-book (MTB-1) ratio and conditional conservatism. Conservatism is
chosen because, even though expensing of ESOs is a conservative accounting treatment, it is not
obvious how the decision would be related to the MTB-1 ratio and conditional conservatism,
particularly because the MTB-1 is a long-run measure of conservatism and the decision to voluntarily
expense is examined over two years. The setting provides a unique opportunity to assess how two
major proxies of conservatism are related.
Design/methodology/approach – Using frms listed in the S&P 1500 index, frms that expensed
ESOs are compared to frms that disclosed only in the fnancial statements. The main comparison uses
a logistic regression.
Findings – During the period when expensing ESOs was voluntary, SFAS 123, the MTB-1 ratio was
negatively associated with ESO expense recognition, but conditional conservatism was positively
associated with ESO expense recognition. The former is attributed to incentives of conservatism and
the latter to the tenet of conservatism tending to reduce income.
Research limitations/implications – The fndings add to the literature/controversy on the
negative relationship between the MTB-1 ratio and conditional conservatism. More important, they
support using more than one measure of conservatismin studies that involve accounting conservatism.
A limitation is that the fndings are specifc to voluntary ESO expense recognition.
Originality/value – This is the frst study to examine how conservatism affects the choice to
recognize an itemon the fnancial statements or disclosure only. In addition, the study shows that frms
were willing to incur real costs from their fnancial reporting.
Keywords Accounting conservatism, Employee stock options, SFAS 123
Paper type Research paper
1. Introduction
This paper investigates the relationship of two accounting conservatism measures, the
beginning-of-period market-to-book (MTB-1) ratio and conditional accounting
The author thanks Pervaiz Alam, Merridee Bujaki, Wei Wei Li, Michael R. Melton, Rafat Noor and
seminar participants at Kent State University. Also, the suggestions andcomments receivedat the
American Accounting Association NE region 2008 conference and Ohio region 2009 conference
are appreciated. This paper was awarded the best PhD student manuscript in 2009 by the AAA
Ohio region.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
ARJ
27,2
124
Accounting Research Journal
Vol. 27 No. 2, 2014
pp. 124-149
©Emerald Group Publishing Limited
1030-9616
DOI 10.1108/ARJ-11-2012-0088
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
conservatism with the choice of whether to recognize employee stock option (ESO)
expense on the fnancial statements or only disclosure in the notes. Conditional
conservatism is proxied by Basu’s (1997) asymmetric timeliness of earnings (AT)
measure. The choice was available due to Statement of Financial Accounting
Standard No. 123 (SFAS 123; Statement of Financial Accounting Standards, 1995),
which up until June 15, 2005, allowed companies to either recognize ESO expense or
to disclose it in the notes to the fnancial statements. This setting provides a unique
opportunity to disentangle how the MTB-1 ratio and conditional conservatism were
related to the conservative treatment of voluntarily expensing ESOs on the fnancial
statements[1].
Examining the MTB-1 ratio, which has been used as a measure of conservatism, and
conditional conservatism is of particular interest because of the controversy in the
literature over how they are related and the need for more research (Pae et al., 2005;
Beatty, 2007; Roychowdhury and Watts, 2007; Nasev, 2009). This study adds to the
literature by examining how the two proxies of conservatism are related to frms’
voluntary disclosure decisions on the fnancial statements; specifcally, the decision to
voluntarily expense ESOs in the fnancial records.
The empirical fndings of this study are consistent with predictions using both
actual and announcing voluntary ESO-expensing frms[2]. Regarding the MTB-1
ratio, frms which voluntarily expensed ESOs during the SFAS 123 period had lower
levels of the MTB-1 ratio. This result supports the incentive and asymmetric
information explanations of conservatism, reducing the need for additional
voluntary information. Regarding conditional conservatism, proxied by Basu’s
(1997) AT, frms which voluntarily expensed ESO during the SFAS 123 period had
higher degrees of conditional conservatism. These results are consistent with
Roychowdhury and Watts’ (2007) term of a “buffer” in net assets from recording
subsequent losses, Beaver and Ryan’s (2005) analytical relationship that shows the
MTB-1 ratio preempts conditional conservatism and Qiang (2007, p. 760) stating
that the MTB-1 ratio “immunizes” the accounting system from future bad news.
Also, conditional conservatism is more of a short-term measure of conservatism.
Further support of both results comes from Pae et al. (2005) and Givoly et al. (2007),
who also show a negative relationship between the MTB-1 ratio and subsequent
conditional conservatism, proxied by AT.
To summarize, this study contributes to the literature on accounting
conservatism and accounting choice in three ways. First, by using a specifc fexible
accounting standard, it shows that both the MTB-1 ratio and conditional
conservatismwere related to a frm’s decision to expense ESOs. Second, the fndings
of this study suggest that during short periods, the MTB-1 ratio and conditional
conservatism can be related in opposite directions, supporting Givoly et al.’s (2007)
views that the two types of conservatism capture different dimensions of
conservatism and that relying on only one measure of conservatism could lead to
erroneous conclusions. Third, this study makes a connection between conservatism
and a frm’s willingness to incur real economic costs from their fnancial reporting
behavior. There were costs to expensing ESOs because many frms lobbied against
expensing stock options, and Guay et al. (2003) discuss real economic costs to be a
possible reason for frms’ opposition.
125
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The remainder of this paper is organized as follows: Section 2 provides background
on ESO accounting standards and presents defnitions and measures of conservatism.
Section 3 develops the empirical predictions, and Section 4 contains the sample selection
process. Section 5 presents the research model, and Section 6 provides the results and
fndings. In Section 7, additional analysis and robustness tests are presented, followed
by conclusions in Section 8.
2. Related prior literature
2.1 Evolution of standards on stock option accounting
The initial accounting standard for stock-based compensation was Accounting
Principles Board Opinion (APB) 25 (1972), Accounting for Stock Issued to Employees,
issued in 1972. This standard permitted companies to account for ESOs using the
“intrinsic value method”, which is the difference between the stock price and the option
exercise price at grant date, without subsequent changes in value. Given that most frms
grant options at an exercise price equal to the current stock price, the ESO expense is
zero using the intrinsic value method (Barth et al., 2012).
Eventually, the Financial Accounting Standards Board (FASB) issued exposure
draft #127-C on June 30, 1993, proposing the requirement of expensing of stock
options using either the Black–Scholes formula or binomial lattice methods for
option pricing at grant date and subsequent adjustments. Many hi-tech frms and
start-up businesses that relied heavily on stock options as an incentive in recruiting
and motivating employees, opposed the move. In contrast, most of the academic
literature agrees that ESOs should be expensed (Botosan and Plumlee, 2001; Bodie
et al., 2003; Guay et al., 2003). For example, Guay et al. (2003, p. 409) conclude:
“Accounting should refect the true costs of doing business, and labor acquired
through ESO grants is a real economic cost that frms should deduct from earnings
as an expense”. Similarly, Bodie et al. (2003) demonstrate that stock option grants
have real cash-fow implications that need to be reported, and detail distortions that
relegating stock option accounting to footnotes creates[3]. Nevertheless, until June
of 2005, frms were allowed to choose whether to recognize the expense or disclose it
in the footnotes of the fnancial statements (SFAS 123, issued October 1995). Almost
all companies decided on disclosure only because until the middle of 2002, just two
companies on the Fortune 500, Boeing and Winn Dixie, had decided to voluntarily
recognize ESO expense. From the middle of 2002 to the frst quarter of 2003, a
signifcant number of companies reported their intention to voluntarily recognize
ESO expense on their fnancial statements. Still, the majority of public companies
chose to only disclose the effects of ESOs in the notes to the fnancial statements
during 2002 through 2004.
SFAS 123(R) was released in December of 2004 and effective for fscal years
beginning June 15, 2005 (see Figure 2 in Section 3). This revision eliminated the option
companies had of only disclosing their ESO expense. From then, all companies, except
small business issuers, were required to use the fair-value method of calculating and
recognizing ESO expense on the fnancial statements as originally prescribed by
Statement 123. This revised standard was developed by the FASBto create a more level
playing feld in terms of management incentive compensation and its impact on a
company’s net income by requiring all forms of compensation to be expensed on the
fnancial statements.
ARJ
27,2
126
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2 Valuation and recognition of ESOs
ESOs have many features that are different to exchange-traded options for which the
Black–Scholes formula was developed (Black and Scholes, 1973). For example, ESOs
usually have a vesting period during which they cannot be exercised, they have
non-transferability, their maturity is much longer (typically ten years) and they can be
exercised before an expiry date. Also, a common reason for giving ESOs is to align
employees’ and executives’ interests with those of the shareholders. Rubinstein
(1995) provides an extensive examination of the various methods and adjustments
needed for valuation of ESOs. He also critiques Exposure Draft 127-C (1993) and
provides additional methods for valuation, which he shows to be better than the
recommendations of the exposure draft.
2.3 Choice in accounting
Fields et al. (2001, p. 256) defne accounting choice as follows:
An accounting choice is any decision whose primary purpose is to infuence (either in form or
substance) the output of the accounting system in a particular way, including not only
fnancial statements published in accordance with GAAP, but also tax returns and regulatory
flings.
They conclude that there is considerable literature on accounting choice. However,
despite improvements in research designs, data sources and computing power, many
questions remain unanswered. The accounting choice dimension this paper investigates
is the voluntary decision to record ESO expense on the fnancial statements or disclose
in the notes to the fnancial statements.
A closely related paper to this study is the one by Hui et al. (2009). They fnd that
conservatism is negatively associated with the issuance of voluntary management
earnings forecasts. However, they do not differentiate between the MTB-1 ratio and
conditional conservatism. More specifc to this study is the paper by Aboody et al.
(2004), who investigate why frms would voluntarily choose to expense options. They
presume that frms would only make this choice if it were cost benefcial and, based on
this premise, identify factors that could explain the decision to expense (see Section 5 for
a discussion).
2.4 Conservatism
The offcial defnition of accounting conservatismis fromFASB Statement of Financial
Accounting Concepts (SFAC) No. 2 (1980)[4]. It describes conservatism as a prudent
reaction to uncertainty and to ensure that uncertainty and risks inherent in a business
situation are adequately considered. For example, if two estimates of amounts to be
received or paid in the future are about equally likely, conservatism requires the less
optimistic estimate to be used. Watts (2003a) and Basu (1997) describe conservatism as
the asymmetry in the verifcation requirements for gains and losses. Firms require a
greater degree of verifcation to record gains than to record losses. Another body of
literature explains conservatismas being induced by economic incentives. In particular,
Watts (2003a, p. 209) argues that conservatism is an “effcient technology employed in
the organization of the frm for its contracts with various parties”. The various parties
are external to the frm and have loss functions that are generally asymmetric. Watts
refers to this as the “contracting explanation” of conservatism and that conservative
accounting tends to reduce contracting costs. Numerous studies document the benefts
127
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
from conservatism in debt contracts (Beatty et al., 2008; Zhang, 2008) and equity
contracts (Ahmed and Duellman, 2007). In these articles, conservatism tends to reduce
income, which provides lenders and investors early signals of default risk, reducing
their contracting costs. As monitoring by external parties is costly, frms have an
incentive to commit to conservatism to get more favorable terms.
Beaver and Ryan (2005) defne unconditional (ex-ante or news-independent)
conservatism as the understatement of book value because of accounting concepts and
standards and conditional conservatism (ex-post or news-dependent) as the
discretionary writing down of assets under adverse conditions[5]. The current paper
interprets unconditional (conditional) conservatism as more of a long (short)-run stock
(fow) type of conservatism that is better represented in the balance sheet (income
statement). Examples of unconditional conservatism include accounting depreciation
that is greater than the actual depreciation of long-term assets, requirements for the
immediate expensing of certain items which last beyond one period and historical
cost accounting that prevents recording future benefts of positive net present value
projects. Jenkins et al. (2009, p. 1044) even state “Unconditional conservatism refers
to the reporting of conservative accounting numbers not conditioned on economic
reality”. Examples of conditional conservatism are how responsive the accounting
system is to new negative information regarding their existing receivables,
inventory and assets/liabilities in general. Figure 1 demonstrates how frms with
different levels of conditional conservatism would adjust their receivables to
information about their probability of collection. The more conditionally
conservative frm is represented by the curved line, which always reports net
receivables at a lower value, ceteris paribus.
2.5 Empirical conservatism measures
Watts (2003b) groups conservatism measures into three types:
(1) earnings and stock returns relation measures;
(2) net asset measures; and
(3) earnings and accrual measures.
100%
Pr. of Default
0 $ Net Receivables
Figure 1.
Conditional conservatism
and recording of
receivables
ARJ
27,2
128
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.5.1 Earnings and stock returns relation measures. Basu (1997) uses a regression of
earnings as a function of positive and negative returns as separate variables. He uses
positive stock returns to generally refect net asset gains and negative stock returns to
refect net asset losses. Basu argues that earnings will refect net losses more quickly
than net gains. This is the AT measure. The AT measure has been criticized as a
measure for conditional conservatism. For example, Givoly et al. (2007) fnd a
signifcantly negative association between the MTB-1 ratio and ATusing periods of 2.5,
5, and 10 years, and Dietrich et al. (2007) fnd biases in the coeffcient estimates of the AT
measure. However, Hsu et al.’s (2012) recent study supports it representing conditional
conservatism. Nevertheless, it is the most commonly used proxy for conditional
conservatism (Ball and Shivakumar, 2005; Beaver and Ryan, 2005; Pae et al., 2005;
Ruddock et al., 2006; Jenkins et al., 2009). Therefore, this study uses AT to proxy for
conditional conservatismand refers to it as a more short-termmeasure than the MTB-1
ratio, which is described in the second point below.
2.5.2 Net asset measures. The use of the MTB-1 ratio (market value divided by the net
book value of equity of the frm) as measure for conservatism emanates from the work
of Felthamand Ohlson (1995), whose defnition of conservatismis based on the disparity
between the market and book values of operating assets. Beaver and Ryan (2005, p. 269)
state, “We defne accountingconservatismas the onaverage understatement of the book
value of net assets relative to their market value (hereafter, the existence of expected
unrecorded goodwill)”. The MTB ratio is also used for conservatism in general. Nasev
(2009, p. 1) notes that the MTB ratio is the most obvious measure of conservatism.
Another study, by Pae et al. (2005), tries to reconcile conservatism in the income
statement with that in the balance sheet. They fnd conditional conservatism, the
tendency of frms to recognize bad news in earnings on a timelier basis than good news
(AT), to be negatively associated with the MTB-1 ratio (beginning-of-period MTBratio).
Roychowdhury and Watts (2007) explain how the correlation between the MTB-1 ratio
and AT can be negative, as observed by Pae et al. (2005), during short horizon periods.
They showthat both the MTB-1 ratio and ATmeasure conservatismwith error, and AT
is commonly not measured over horizons that include a frm’s initial public offering.
They demonstrate a positive association between the MTB-1 and AT over long time
horizons.
2.5.3 Accrual measures. Mason (2004) constructs a conservatismindex by assigning
values to the amount of R&D, advertising, inventory, unrecorded pension, accounts
receivable and depreciable assets. These are Mason’s proxies for the unrecorded
(accrual) value of a frm’s operating assets. Penman and Zhang (2002) also develop a
conservatismindex. They defne Ci ?ERi/NOAi, where Ci is the conservatismindex for
frmi, ER is the estimated reserve and NOAis the net operating assets. The ER is based
on the capitalization of R&D, advertising and differences in inventory between LIFO
and FIFO accounting policies. A complete conservatism index would include all
operating reserves, but the authors only choose these items because they are not subject
to management discretion.
3. Hypotheses development
The voluntary expensing of ESOs is consistent with the defnition(s) of conservatism
described in the literature review, and both the MTB-1 ratio and conditional
129
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
conservatism, proxied by AT, have been used to represent conservatism. Nevertheless,
the MTB-1 ratio and conditional conservatism represent different dimensions of
conservatism.
First, the MTB-1 ratio is the understatement of book values to their respective market
values, and a higher MTB-1 ratio implies greater past conservatism, ceteris paribus.
Given that expensing ESOs is a conservative accounting treatment, this supports a
positive relationship between expensing and the MTB-1 ratio. However, this positive
relationship is weakened because the MTB-1 ratio is a long-run measure of
conservatism that has been developing since the beginning of a company’s life. The
voluntary expensing of ESOs is only examined over a short period, 2003 and 2004, so a
frm’s short-run degree can be different from its long-run degree of conservatism (i.e.
MTB-1 ratio). This is even more likely for frms that have an increase in their market
value due to an increase in unrecorded intangibles that are unrelated to their degree
of conservatism. Further, frms that decided to voluntarily expense ESOs could have
had an incentive to do so because they have not been conservative in the past (i.e.
low MTB-1 ratio). This reasoning is supported by the contracting explanation of
conservatism whereby conservatism reduces information asymmetry to users
outside the frm (Watts, 2003a; LaFond and Watts, 2008). Furthermore, Gigler and
Hemmer (2001) develop theoretical links about the relationship between
conservatism and timely voluntary disclosures using a principal-agent model. They
fnd that less conservative frms would make more timely disclosures and thereby
provide greater voluntary information to the market. Similarly, Hui et al. (2009) fnd
a negative association between the MTB ratio and the issuance of voluntary
management earnings forecasts. They suggest that conservatism acts as a
substitute for management forecasts making conservative frms have less incentive
to voluntarily issue management forecasts.
Therefore, given the above arguments, the MTB-1 ratio’s relationship to the decision
to voluntarily expense ESOs is not obvious and therefore an empirical question. The
frst hypothesis for the MTB-1 ratio is presented in non-directional form:
H1. Ceteris paribus, during the period when expensing was not mandatory (SFAS
123), the beginning-of-period MTB-1 ratio is related to frms’ decision to
voluntarily expense stock options.
The second hypothesis is how conditional conservatism is related to the voluntary
expensing of ESOs. Conditional conservatism can be described as the recording of
income-decreasing economic events on a timelier basis than income-increasing
economic events. Clearly, the decision to expense ESOs is income-decreasing, which
supports a positive relationship. However, as discussed in formulating H1,
incentives also play a role in voluntary accounting choices. It was argued that frms
would have more incentive to expense if they had a low MTB-1 ratio. If this holds
true, then it adds more support for conditional conservatismbeing positively related
to expensing. Beaver and Ryan (2005) and Lawrence et al. (2012) fnd that
conditional conservatism is higher when frms have low MTB ratios. Therefore, a
positive relationship between the degree of a frm’s conditional accounting
conservatism, proxied by AT, and voluntary expensing of ESOs is predicted. The
second hypothesis is presented in directional form:
ARJ
27,2
130
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
H2. Ceteris paribus, during the period when expensing was not mandatory (SFAS
123), conditional accounting conservatismis positively related to frms’ decision
to voluntarily expense stock options.
Finally, although not a formal hypothesis, higher stock return volatility of a frm is
predicted to make a frm less likely to expense ESOs in favor of disclosure only on the
fnancial statements. This prediction is based on the assumption that frms would not
want to show a large amount for ESO expense, and large variations in comparative net
income may be interpreted as higher idiosyncratic risk for the frm[6]. In any case, not
including omitting this could lead to an omitted variable bias.
4. Data sources and periods
All available US frm-level fnancial data were obtained from Standard & Poor (S&P)’s
Research Insight Compustat database. However, the sample is restricted to frms in the
S&P 500, S&P 400 mid-capitalization and S&P 600 small-capitalization indices. This
group is selected because most of the ESO expense-recognizing frms are included in
these indices and covered by the ExecuComp database. Stock return and market
valuation data are obtained from the CRSP (Center for Research on Security Prices)
databases, and information regarding stock-based compensation and Black–Scholes
volatility is obtained from S&P’s ExecuComp database. For companies with missing
volatility numbers in ExecuComp, the volatility is calculated as the standard deviation
of annual returns over the past fve years. In addition, CEO data for ESO expensing
companies missing in ExecuComp are hand-collected from the Defnitive Notice &
Proxy Statement Schedule 14A.
The year selected for a frmto have voluntarily chosen to expense ESOs is fscal year
2004, fromwhich 155 (547) frms in the S&P 1500 index (Compustat population) had OE
in the Compustat Mnemonic STKCOF. OE in this Mnemonic signifes that the frmwas
an early adaptor of SFAS 123R. However, pooled cross-sectional testing is performed
over two fscal years – 2003 and 2004 – to allow for enough ESO-expensing frms to
obtain meaningful results from the conservatism proxies[7][8]. Therefore, the same
frms that expensed ESOs in fscal year 2004 are used for 2003. Figure 2 presents a
timeline that shows the various dates used for testing the hypotheses in this study and
for SFAS 123 and SFAS 123(R).
5. Research design
To test the hypotheses, frms that expensed ESOs are compared to frms that disclosed
only in the footnotes of the fnancial statements for the fscal year 2004. The main
comparison uses a logit regression of pooled cross-sectional frm-year observations for
fscal years 2003 and 2004. This is accomplished by using the same frms that expensed
ESOs in 2004 for years 2003 and 2004. The comparison group are frms in the S&P 1500
that did not expense ESOs.
The model used in this paper borrows heavily from the logit model of Aboody et al.
(2004), which examines likely factors associated with frms that voluntarily announced
Beg. fiscal yr. 2003 Beg. fiscal yr. 2004 End fiscal yr. 2004 June 15, 2005
Inclusion Year SFAS 123R
Hypotheses testing period required
Figure 2.
Timeline of SFAS 123 and
testing period
131
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that they would expense ESOs[9]. They categorize their independent variables into four
areas that are likely associated with expensing ESOs: capital market activity, private
incentives of management and the board of directors, political costs of not expensing
ESOs and controlling for the magnitude of SFAS 123 expense. The current paper
incorporates similar variables, including their predicted direction, in the model of this
study but also includes the MTB-1 ratio, conditional conservatism(AT) and stock return
volatility.
The model is presented below as equation (1):
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB ? 1, AT, SKEW
?
it ? ?
it
(1)
The dependent variable, EXPD, is a dichotomous variable that equals 1 for an ESO
expense-recognizing frm (Compustat Mnemonic STKCOF ? OE) and 0 for
disclosure-only frms for fscal year 2004. Next, IND
N
are indicator variables that equal
1 if the frmis in industry N, and 0 otherwise, based on SIC code classifcations grouped
into ten categories. The industry variables are to control for the potential effects of
omitted variables associated with industry membership[10].
The frst set of variables relate to capital market activity, which are predicted to be
positively related to SFAS 123 ESO expense recognition. ISSUEp is the percentage
change in common shares outstanding (Compustat #A25); AQUp is the ratio of cash
outfow of funds used for and/or the costs relating to acquisition of a company
(Compustat #A129) defated by total assets (COMPUSTAT #6); DEBTEQ
(Compustat Mnemonic DCE) is the ratio of long-term debt to common equity; and
INTCOV is the ratio of interest expense (Compustat #A15) to net operating income
(Compustat #A178).
The second set relates to private incentives of management. BONUS is the ratio of
CEO cash bonus to total cash compensation and predicted to be negatively related to
ESO expense recognition; CEOOWN is equity shares and options held by the CEO as a
percentage of total shares outstanding and predicted to be positively related to ESO
expense recognition[11]. In between the above set of variables and the ones in the next
paragraph is the percentage of shares outstanding held by institutional investors
(INSTp) used as a proxy for information asymmetry. It is predicted that the greater the
institutional investor percentage, the less likely the frm will expense ESOs because
institutional investors are more sophisticated, which enables themto better understand
footnote disclosure.
The next sets of variables are characterized as political costs of not expensing
ESOs and are predicted to have a positive relation with voluntary ESO expense
recognition. SIZE is the logarithm of market value of equity (Compustat Mnemonic
MKVALF); PROFITp is net income (Compustat #A172) defated by market value of
equity.
Consistent with the Aboody et al. (2004) model, the magnitude of SFAS 123 expense
is controlled by including NI_PF, which is the difference between reported net income
ARJ
27,2
132
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and S&P core earnings (Compustat Mnemonic COREEARN), defated by market value
of equity. Finally, the interaction of NI_PF and INSTp is also included to control for
information asymmetry.
BSVOL is the Black–Scholes volatility measure obtained from the ExecuComp
database[12]. This variable is important in the expense recognition decision because a
highly volatile stock would increase the value and variability of ESOs, which may
discourage a company from expensing ESOs if they are concerned that this would
signifcantly affect their net income and its variability. Therefore, this variable is
predicted to be negatively related to ESO expense recognition. The fnal variables are
the conservatism proxies, which will be substituted individually by the
beginning-of-period MTB-1 ratio (Compustat Mnemonic MKBKF), AT measure for
conditional conservatism and skewness of cash fows to earnings (SKEW). The MTB-1
ratio attempts to capture conservatism that has been developing since the inception of
the frm. AT is calculated using the procedure described in the next paragraph and
SKEWis another proxy for conservatismwhich is discussed in Section 7.2 as additional
analysis.
Basu (1997) measured a frm’s degree of conservatismas the sensitivity of earnings to
negative returns. In this paper, Basu’s model of earnings as a function of
contemporaneous stock returns [equation (2)] is used for the measure of conditional
conservatism (AT).
X
it
/P
it?1
? ? ? ?
1
DR
it
? ?
2
R
it
? ?
3
(R
it
xDR
it
) ? ?
it
(2)
Where:
X
it
?Earnings per share before extra-ordinary items and discontinued operations
[Compustat item #58];
R
it
?annual cum dividend return from CRSP;
DR
it
? a dummy variable which equals 1, when returns are negative, and 0
otherwise;
P
it-1
?end of prior fscal year stock price [Compustat item #199].
The main coeffcient of interest is ?
3
, which represents the incremental effect on
earnings from negative returns (coined “bad news” in Basu’s article), in comparison to
positive returns (good news). Apositive and statistically signifcant ?
3
is interpreted as
existence of accounting conservatism because negative returns have a greater effect on
earnings than do positive returns. Following Francis et al. (2004) and Zhang (2008), the
sensitivity of earnings to negative returns (?
2
??
3
), relative to positive returns (?
2
), is
used as the proxy for frm-level conditional conservatism – AT ?(?
2,it
??
3,it
/?
2,it
)[13].
This quotient is calculated on a rolling ten-year window, t-9 […], t. A minimum fve
years of earnings and returns is required for a frm-year observation to be given an AT
measure of conservatism. It is expected that AT will be greater than one.
6. Empirical results
6.1 Descriptive statistics
Table I, panels Aand B, report the breakdown of the variables used in this paper. Panel
Apresents descriptive statistics for the whole sample and panel B has frms partitioned
by recognizing ESOs on the fnancial statements and disclosure-only frms. All
133
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table I.
Descriptive statistics for
conservatism proxies,
return volatility and
control variables
P
a
n
e
l
A
S
a
m
p
l
e
p
e
r
i
o
d
:
2
0
0
3
-
2
0
0
4
V
a
r
i
a
b
l
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
M
e
d
i
a
n
2
5
t
h
p
e
r
c
e
n
t
i
l
e
7
5
t
h
p
e
r
c
e
n
t
i
l
e
F
Y
R
1
,
7
7
4
2
,
0
0
3
.
5
0
.
5
0
0
2
,
0
0
4
.
0
2
,
0
0
3
.
0
2
,
0
0
4
.
0
I
S
S
U
E
p
1
,
7
7
4
0
.
0
2
2
0
.
0
6
4
0
.
0
1
0
0
.
0
0
0
0
.
0
2
6
A
Q
U
p
1
,
7
7
4
0
.
0
2
4
0
.
0
5
5
0
.
0
0
0
0
.
0
0
0
0
.
0
1
8
D
E
B
T
E
Q
1
,
7
7
4
0
.
6
2
0
0
.
8
7
2
0
.
4
0
4
0
.
1
0
1
0
.
8
0
5
I
N
T
C
O
V
1
,
7
7
4
0
.
1
9
6
0
.
3
8
1
0
.
1
0
9
0
.
0
3
0
0
.
2
6
3
B
O
N
U
S
1
,
7
7
4
0
.
4
4
3
0
.
2
4
4
0
.
4
9
5
0
.
3
1
2
0
.
6
1
9
C
E
O
O
W
N
1
,
7
7
4
0
.
0
3
5
0
.
0
5
6
0
.
0
1
6
0
.
0
0
8
0
.
0
3
5
I
N
S
T
p
1
,
7
7
4
0
.
8
6
5
0
.
1
6
0
0
.
8
8
7
0
.
7
7
2
0
.
9
6
9
S
I
Z
E
1
,
7
7
4
8
,
2
0
4
1
9
,
0
7
6
2
,
0
1
0
8
4
8
6
,
0
3
9
P
R
O
F
I
T
p
1
,
7
7
4
0
.
0
4
0
0
.
0
5
6
0
.
0
4
7
0
.
0
3
1
0
.
0
6
2
N
I
_
P
F
1
,
7
7
4
0
.
0
0
4
0
.
0
2
6
0
.
0
0
4
0
.
0
0
1
0
.
0
0
9
B
S
V
O
L
1
,
7
7
4
0
.
4
3
7
0
.
2
0
3
0
.
3
8
3
0
.
2
9
8
0
.
5
3
4
M
T
B
-
1
1
,
7
7
1
2
.
7
2
3
2
.
3
2
4
2
.
0
9
8
1
.
4
7
8
3
.
2
6
8
A
T
1
,
4
6
4
2
.
1
5
1
3
.
2
4
4
0
.
4
0
6
0
.
0
0
0
2
.
8
4
3
S
S
K
E
W
1
,
7
4
3
0
.
2
8
5
0
.
2
4
4
0
.
2
5
6
0
.
1
4
7
0
.
3
8
0
P
a
n
e
l
B
V
a
r
i
a
b
l
e
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
D
i
s
c
l
o
s
u
r
e
-
o
n
l
y
f
r
m
s
D
i
f
f
e
r
e
n
c
e
i
n
m
e
a
n
s
t
-
s
t
a
t
p
-
v
a
l
u
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
I
S
S
U
E
p
2
4
0
0
.
0
2
8
0
.
0
7
0
1
,
5
3
4
0
.
0
2
1
0
.
0
6
2
0
.
0
0
7
1
.
5
6
0
0
.
1
1
9
A
Q
U
p
2
4
0
0
.
0
0
8
0
.
0
2
7
1
5
3
4
0
.
0
2
7
0
.
0
5
8
?
0
.
0
1
8
?
4
.
8
3
0
?
.
0
0
0
1
D
E
B
T
E
Q
2
4
0
0
.
8
2
5
0
.
9
4
0
1
5
3
4
0
.
5
8
8
0
.
8
5
6
0
.
2
3
7
3
.
9
3
0
?
.
0
0
0
1
I
N
T
C
O
V
2
4
0
0
.
2
5
2
0
.
3
4
5
1
5
3
4
0
.
1
8
8
0
.
3
8
6
0
.
0
6
4
2
.
4
4
0
0
.
0
1
5
B
O
N
U
S
2
4
0
0
.
5
2
0
0
.
2
5
2
1
5
3
4
0
.
4
3
1
0
.
2
4
0
0
.
0
8
9
5
.
3
0
0
?
.
0
0
0
1
C
E
O
O
W
N
2
4
0
0
.
0
2
4
0
.
0
4
5
1
5
3
4
0
.
0
3
7
0
.
0
5
7
?
0
.
0
1
3
?
3
.
4
5
0
?
.
0
0
0
1
I
N
S
T
p
2
4
0
0
.
8
2
9
0
.
1
7
5
1
5
3
4
0
.
8
7
0
0
.
1
5
6
?
0
.
0
4
1
?
3
.
7
1
0
?
.
0
0
0
1
(
c
o
n
t
i
n
u
e
d
)
ARJ
27,2
134
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table I.
P
a
n
e
l
B
V
a
r
i
a
b
l
e
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
D
i
s
c
l
o
s
u
r
e
-
o
n
l
y
f
r
m
s
D
i
f
f
e
r
e
n
c
e
i
n
m
e
a
n
s
t
-
s
t
a
t
p
-
v
a
l
u
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
S
I
Z
E
2
4
0
1
6
,
7
9
6
3
1
,
1
9
7
1
5
3
4
6
,
8
6
0
1
5
,
9
9
3
9
,
9
3
6
7
.
6
2
0
?
.
0
0
0
1
P
R
O
F
I
T
p
2
4
0
0
.
0
5
4
0
.
0
4
7
1
5
3
4
0
.
0
3
7
0
.
0
5
7
0
.
0
1
6
4
.
2
9
0
?
.
0
0
0
1
N
I
_
P
F
2
4
0
0
.
0
0
4
0
.
0
2
9
1
5
3
4
0
.
0
0
4
0
.
0
2
5
0
.
0
0
0
0
.
1
4
0
0
.
8
8
5
B
S
V
O
L
2
4
0
0
.
3
2
5
0
.
1
5
7
1
5
3
4
0
.
4
5
5
0
.
2
0
3
?
0
.
1
3
0
?
9
.
4
5
0
?
.
0
0
0
1
M
T
B
-
1
2
4
0
2
.
3
6
4
2
.
0
2
4
1
5
3
1
2
.
7
7
9
2
.
3
6
3
?
0
.
4
1
5
2
.
5
7
0
0
.
0
1
0
A
T
1
8
9
2
.
6
9
1
3
.
5
6
9
1
2
7
5
2
.
0
7
1
3
.
1
8
7
0
.
6
2
0
2
.
4
6
0
0
.
0
1
4
S
S
K
E
W
2
3
5
0
.
2
1
8
0
.
2
0
3
1
5
0
8
0
.
2
9
5
0
.
2
4
8
?
0
.
0
7
7
?
4
.
5
5
0
?
.
0
0
0
1
N
o
t
e
s
:
I
S
S
U
E
p
i
s
t
h
e
p
e
r
c
e
n
t
a
g
e
c
h
a
n
g
e
i
n
c
o
m
m
o
n
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
(
#
A
2
5
)
;
A
Q
U
p
i
s
t
h
e
r
a
t
i
o
o
f
c
a
s
h
o
u
t
f
o
w
o
f
f
u
n
d
s
u
s
e
d
f
o
r
a
n
d
/
o
r
c
o
s
t
s
r
e
l
a
t
i
n
g
t
o
a
c
q
u
i
s
i
t
i
o
n
o
f
a
c
o
m
p
a
n
y
(
#
A
1
2
9
)
d
e
f
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
(
#
6
)
;
D
E
B
T
E
Q
(
M
n
e
m
o
n
i
c
D
C
E
)
i
s
t
h
e
r
a
t
i
o
o
f
l
o
n
g
-
t
e
r
m
d
e
b
t
t
o
c
o
m
m
o
n
e
q
u
i
t
y
;
I
N
T
C
O
V
i
s
t
h
e
r
a
t
i
o
o
f
i
n
t
e
r
e
s
t
e
x
p
e
n
s
e
(
#
A
1
5
)
t
o
n
e
t
o
p
e
r
a
t
i
n
g
i
n
c
o
m
e
(
#
A
1
7
8
)
;
B
O
N
U
S
i
s
t
h
e
r
a
t
i
o
o
f
C
E
O
c
a
s
h
b
o
n
u
s
t
o
t
o
t
a
l
c
a
s
h
c
o
m
p
e
n
s
a
t
i
o
n
;
C
E
O
O
W
N
i
s
e
q
u
i
t
y
s
h
a
r
e
s
a
n
d
o
p
t
i
o
n
s
h
e
l
d
b
y
t
h
e
C
E
O
a
s
a
p
e
r
c
e
n
t
a
g
e
o
f
t
o
t
a
l
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
;
I
N
S
T
p
i
s
t
h
e
p
e
r
c
e
n
t
a
g
e
o
f
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
;
S
I
Z
E
i
s
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
(
M
n
e
m
o
n
i
c
M
K
V
A
L
F
)
;
P
R
O
F
I
T
p
i
s
n
e
t
i
n
c
o
m
e
(
#
A
1
7
2
)
d
e
f
a
t
e
d
b
y
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
;
N
I
_
P
F
i
s
t
h
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
r
e
p
o
r
t
e
d
n
e
t
i
n
c
o
m
e
a
n
d
S
&
P
c
o
r
e
e
a
r
n
i
n
g
s
(
M
n
e
m
o
n
i
c
C
O
R
E
E
A
R
N
)
d
e
f
a
t
e
d
b
y
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
;
B
S
V
O
L
i
s
t
h
e
B
l
a
c
k
–
S
c
h
o
l
e
s
v
o
l
a
t
i
l
i
t
y
m
e
a
s
u
r
e
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
E
x
e
c
u
C
o
m
p
d
a
t
a
b
a
s
e
;
M
T
B
-
1
i
s
t
h
e
b
e
g
i
n
n
i
n
g
-
o
f
-
p
e
r
i
o
d
m
a
r
k
e
t
-
t
o
-
b
o
o
k
r
a
t
i
o
(
M
n
e
m
o
n
i
c
M
K
B
K
F
)
;
A
T
i
s
t
h
e
s
e
n
s
i
t
i
v
i
t
y
o
f
e
a
r
n
i
n
g
s
t
o
n
e
g
a
t
i
v
e
r
e
t
u
r
n
s
c
a
l
c
u
l
a
t
e
d
o
v
e
r
1
0
y
e
a
r
s
(
m
i
n
i
m
u
m
o
f
5
y
e
a
r
s
)
;
a
n
d
S
K
E
W
i
s
t
h
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
c
a
s
h
f
o
w
s
(
#
3
0
8
)
a
n
d
n
e
t
i
n
c
o
m
e
(
#
1
7
2
)
d
e
f
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
(
#
6
)
s
u
m
m
e
d
o
v
e
r
t
h
e
p
r
i
o
r
f
v
e
f
s
c
a
l
y
e
a
r
s
;
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
o
f
2
4
0
a
r
e
a
f
t
e
r
r
e
q
u
i
r
i
n
g
a
l
l
v
a
r
i
a
b
l
e
s
t
o
h
a
v
e
a
v
a
l
u
e
,
o
t
h
e
r
t
h
a
n
t
h
e
c
o
n
s
e
r
v
a
t
i
s
m
p
r
o
x
i
e
s
;
L
o
g
a
r
i
t
h
m
o
f
S
I
Z
E
i
s
u
s
e
d
i
n
r
e
g
r
e
s
s
i
o
n
s
.
A
l
l
c
o
n
t
i
n
u
o
u
s
v
a
r
i
a
b
l
e
s
a
r
e
w
i
n
s
o
r
i
z
e
d
a
t
1
a
n
d
9
9
p
e
r
c
e
n
t
o
f
t
h
e
i
r
r
e
s
p
e
c
t
i
v
e
d
i
s
t
r
i
b
u
t
i
o
n
s
135
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
continuous independent variables are winsorized at the top and bottom 1 per cent of
their respective distributions to mitigate the infuence of outliers.
The two main conservatism proxies examined in this study are the
beginning-of-period MTB-1 ratio and the AT, the latter representing conditional
conservatism. Panel Bshows the MTB-1 ratio of expensing compared to disclosure-only
frms is lower by 0.415 (18 per cent), and the AT measure of expensing compared to
disclosure-only frms is higher by 0.620 (23 per cent). Also shown in Panel B is the
SKEW measure of conservatism. It is lower for the ESO expense-recognizing frms and
is further discussed in Section 7.2. With respect to the control variables, differences
include a lower ACUp, higher DEBTEQ, higher INTCOV, higher BONUS, lower
CEOOWN, higher SIZE and higher PROFITp for the ESO-expensing frms. ISSUEp,
INSTp and NI_PF are similar between the two groups[14].
This study also examines how stock volatility was related to voluntarily expensing
ESOs. This variable, BSVOL, is 0.130 (29 per cent) lower for expensing frms, suggesting
that higher stock return volatility reduced the likelihood of ESO expense recognition.
Table II presents the industry classifcation for recognizing and disclosure-only
frms. A major difference between recognizing and disclosure-only frms is in the
fnancial services industry. Financial services constitute approximately 40 per cent of
Table II.
Industry classifcation
and descriptive statistics
Sample period: 2003-2004 Recognizing frms Disclosure-only frms
Industry Number (%) Number (%)
Other 2 0.83 11 0.72
1 Chemicals 9 3.75 57 3.72
2 Computers 3 1.25 193 12.58
3 Extractive 14 5.83 66 4.30
4 Financial 50 20.83 42 2.74
5 Food 8 3.33 86 5.61
6 Insurance/Real Estate 47 19.58 23 1.50
7 Manf: Electrical equipment 2 0.83 51 3.32
8 Manf: Instruments 0 0.00 108 7.04
9 Manf: Machinery 3 1.25 77 5.02
10 Manf: Metal 6 2.50 52 3.39
11 Manf: Misc. 0 0.00 19 1.24
12 Manf: Rubber/glass/etc. 2 0.83 27 1.76
13 Manf: Transport equipment 6 2.50 42 2.74
14 Mining/Construction 8 3.33 38 2.48
15 Pharmaceuticals 2 0.83 53 3.46
16 Retail: Misc. 15 6.25 111 7.24
17 Retail: Restaurant 2 0.83 29 1.89
18 Retail: Wholesale 10 4.17 59 3.85
19 Services 11 4.58 122 7.95
20 Textiles/Print/Publ 6 2.50 87 5.67
21 Transportation 10 4.17 66 4.30
22 Utilities 24 10.00 115 7.50
Totals 2 40 1 00
Notes: The above industries are summarized fromfour-digit SIC codes into 22 industries; Category 4
and 6, in italics, are considered fnancial services companies
ARJ
27,2
136
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the recognizing frms but only 4 per cent of the disclosure-only frms. Additional testing
that excludes fnancial services frms is performed in Section 7.3.
6.2 Univariate analysis of the association between variables
Table III presents the Pearson and Spearman correlation coeffcients among all the
variables used in the model. The univariate relationships of whether a frm expensed
ESOs on the fnancial statements (EXPD) with the independent variables are generally
in the same direction as the multivariate results reported in the next section. Of more
interest to this study is how EXPD is related to the conservatism proxies. EXPD is
negatively related to the MTB-1 ratio, but positively related to the conditional
conservatism proxy (AT) with correlation coeffcients (?) of ?0.088 and 0.064,
respectively. In addition, correlation coeffcient between the MTB-1 ratio and AT is
negative with a (?) of ?0.035 with a p value of 0.178, suggesting that they are negatively
related for the current sample. Nevertheless, it is diffcult to draw defnitive
implications, as the univariate analysis ignores the potential impact of correlations
between other explanatory variables. Multivariate analysis is required to address the
main questions of this paper, which is mentioned in the next sub-section.
6.3 Analysis of logistic regression results
The results of the logistic model [equation (1)] for fscal years 2003 and 2004 are shown
in Table IV (Columns 1, 2 and 3). Column 1 presents the coeffcients (Wald chi-square)
for the variables used from the Aboody et al. (2004) model without the conservatism
proxies[15]. The main coeffcients of interest of this paper are fromvariables MTB-1 and
AT, in Table IV, Columns 2 and 3, respectively. First, the coeffcient on MTB-1 is
negatively related to whether a frm expensed ESOs with a value of ?0.082 and
statistically signifcant at the 10 per cent level. The hypothesis of the relationship
between EXPD and the MTB-1 ratio, proxied by MTB-1, was presented as
non-directional because the income-decreasing tendency of conservatismand incentives
to expense ESOs were in opposite directions. In any case, H1 is supported because the
MTB-1 ratio is related to EXPDat a statistically signifcant level. Ahigher MTB-1 ratio
is interpreted in this paper as being relatively more conservative in the past. Given that
the coeffcient on MTB-1 is negative suggests that frms which had higher MTB-1 ratios
had less of an incentive to expense ESOs. Conservatism is documented to reduce
information asymmetry, so it is possible that expensing ESOs would not materially help
to reduce information asymmetry for high MTB-1 ratio frms.
Second, the coeffcient on AT is positively related to a frm expensing ESOs with a
value of 0.068 and statistically signifcant at the 1 per cent level (Column 3), indicating
that voluntary expensing of ESOs was more likely associated with frms that were
conditionally conservative, supporting H2[16]. The arguments for conditional
conservatism being positively related to the likelihood of ESO expense recognition
rested mainly on conditional conservatism’s asymmetric requirements to record
income-reducing items in a timelier manner than income-increasing items. Expensing
ESOs will lower income compared to disclosure only. In addition, the hypothesis
development for H2 in Section 3 did not predict conditional conservatismto be related to
incentives for expensing ESOs.
The above two results show that the MTB-1 ratio is negatively related and AT
positively related to the voluntary expensing of ESOs. Apossible explanation for this is
137
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table III.
Correlation analysis
among dependent and
independent variables
Pearson
(above)/Spearman (below)
V
a
r
i
a
b
l
e
E
X
P
D
I
S
S
U
E
p
A
Q
U
p
D
E
B
T
E
Q
I
N
T
C
O
V
B
O
N
U
S
C
E
O
O
W
N
I
N
S
T
p
S
I
Z
E
P
R
O
F
I
T
p
N
I
_
P
F
B
S
V
O
L
M
T
B
-
1
A
T
S
K
E
W
E
X
P
D
0
.
0
3
7
?
0
.
1
1
4
0
.
1
0
1
?
0
.
0
6
1
0
.
1
2
5
?
0
.
0
8
2
0
.
0
9
3
0
.
1
7
8
0
.
0
0
3
0
.
0
5
8
?
0
.
2
1
9
?
0
.
0
8
8
0
.
0
6
4
?
0
.
1
0
8
I
S
S
U
E
p
?
0
.
0
0
8
0
.
1
6
0
?
0
.
1
1
6
?
0
.
1
1
1
0
.
0
0
1
?
0
.
0
4
1
0
.
0
6
7
?
0
.
0
9
9
0
.
0
0
3
0
.
1
0
4
0
.
0
7
4
0
.
0
3
8
?
0
.
0
0
5
0
.
0
3
3
A
Q
U
p
?
0
.
1
2
3
0
.
0
2
6
?
0
.
0
0
4
0
.
0
3
3
0
.
0
2
5
0
.
0
3
5
0
.
0
0
1
?
0
.
0
6
4
0
.
0
0
0
?
0
.
0
8
8
0
.
0
1
7
0
.
0
9
5
?
0
.
0
3
9
0
.
0
3
5
D
E
B
T
E
Q
0
.
1
5
1
?
0
.
1
1
6
?
0
.
0
2
0
0
.
0
5
5
0
.
2
9
3
0
.
0
3
1
?
0
.
1
4
5
0
.
0
5
4
0
.
5
3
3
?
0
.
1
6
1
?
0
.
3
1
5
?
0
.
0
3
0
0
.
0
2
2
?
0
.
1
7
8
I
N
T
C
O
V
?
0
.
0
5
7
?
0
.
1
7
6
0
.
1
1
1
?
0
.
0
7
3
0
.
1
3
6
?
0
.
0
1
0
0
.
0
0
5
0
.
2
4
8
0
.
0
4
8
?
0
.
1
9
7
?
0
.
0
5
1
?
0
.
0
5
6
?
0
.
0
0
6
0
.
0
3
0
B
O
N
U
S
0
.
1
4
1
?
0
.
0
3
1
0
.
0
7
3
0
.
3
1
0
0
.
1
3
9
?
0
.
1
2
0
0
.
0
3
7
0
.
2
4
7
0
.
0
1
0
?
0
.
0
3
9
?
0
.
2
0
3
?
0
.
0
4
2
0
.
0
0
3
?
0
.
0
9
7
C
E
O
O
W
N
?
0
.
1
9
1
0
.
0
3
4
0
.
0
6
7
?
0
.
0
6
1
?
0
.
0
0
8
?
0
.
1
4
8
?
0
.
1
2
8
?
0
.
1
2
5
0
.
0
6
4
?
0
.
0
5
6
0
.
1
1
8
?
0
.
0
9
7
?
0
.
0
4
7
?
0
.
0
1
4
I
N
S
T
p
0
.
1
2
4
0
.
0
9
6
0
.
0
1
0
0
.
1
3
7
?
0
.
1
8
5
0
.
0
7
0
?
0
.
2
1
6
?
0
.
0
1
5
?
0
.
1
4
0
0
.
3
5
1
?
0
.
0
9
3
?
0
.
0
3
5
0
.
0
2
0
?
0
.
1
2
7
S
I
Z
E
0
.
1
5
5
?
0
.
1
9
6
0
.
0
1
9
0
.
1
0
9
0
.
3
2
7
0
.
4
3
6
?
0
.
4
3
2
0
.
1
2
2
?
0
.
0
0
6
?
0
.
0
6
0
?
0
.
1
6
3
?
0
.
2
9
8
0
.
0
2
1
?
0
.
0
2
7
P
R
O
F
I
T
p
?
0
.
0
7
2
0
.
0
9
4
?
0
.
0
5
0
0
.
0
6
3
?
0
.
0
2
1
?
0
.
1
0
5
0
.
1
6
3
?
0
.
0
7
8
?
0
.
1
3
8
?
0
.
1
0
4
0
.
0
1
6
0
.
0
2
4
0
.
0
0
8
0
.
0
2
7
N
I
_
P
F
0
.
1
0
9
0
.
1
7
3
?
0
.
0
7
7
0
.
0
9
3
?
0
.
3
9
5
?
0
.
0
0
3
?
0
.
1
9
2
0
.
7
2
3
0
.
0
0
2
?
0
.
0
3
0
?
0
.
0
1
7
?
0
.
0
1
1
?
0
.
0
1
4
?
0
.
0
5
2
B
S
V
O
L
?
0
.
2
6
3
0
.
1
2
0
?
0
.
0
1
5
?
0
.
3
4
7
?
0
.
0
5
1
?
0
.
1
9
6
0
.
2
8
2
?
0
.
2
8
7
?
0
.
3
1
3
0
.
1
9
8
?
0
.
2
0
9
0
.
1
8
4
0
.
0
4
9
0
.
3
2
6
M
T
B
-
1
?
0
.
0
8
0
0
.
1
2
0
0
.
0
7
6
?
0
.
0
8
5
?
0
.
0
3
9
?
0
.
0
7
7
0
.
1
6
0
?
0
.
0
4
4
?
0
.
2
4
7
0
.
1
3
4
?
0
.
0
2
8
0
.
2
2
9
?
0
.
0
3
5
0
.
0
8
8
A
T
0
.
0
5
1
?
0
.
0
3
2
0
.
0
0
4
0
.
0
3
1
0
.
0
5
6
0
.
0
3
1
?
0
.
0
2
1
?
0
.
0
4
4
0
.
0
3
4
?
0
.
0
5
8
?
0
.
0
6
2
0
.
0
3
7
?
0
.
0
5
9
0
.
0
8
7
S
K
E
W
?
0
.
1
3
2
?
0
.
0
0
9
0
.
0
4
3
?
0
.
1
6
1
0
.
1
0
9
?
0
.
0
6
6
0
.
0
4
2
?
0
.
1
2
3
?
0
.
0
6
6
0
.
0
2
7
?
0
.
1
2
2
0
.
2
2
1
0
.
0
8
8
0
.
0
6
0
N
o
t
e
s
:
A
l
l
v
a
r
i
a
b
l
e
s
a
r
e
a
s
d
e
s
c
r
i
b
e
d
i
n
T
a
b
l
e
I
;
c
o
r
r
e
l
a
t
i
o
n
s
i
n
i
t
a
l
i
c
s
a
r
e
n
o
t
s
i
g
n
i
f
c
a
n
t
a
t
t
h
e
1
0
p
e
r
c
e
n
t
l
e
v
e
l
ARJ
27,2
138
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that because ESO-expensing frms tended to have lower MTB-1 ratios, they had a
“buffer” in net assets for recording subsequent losses, which allows a frm greater
ability to write down assets. This indirectly increases AT by AT’s mechanical
connection to MTB-1. Another explanation is that frms with low MTB-1 ratios are
likely to get more pressure fromauditors to write down their assets (Kinney and Uecker,
1982; Pae et al., 2005), again indirectly increasing AT. Further, these results add to the
observed negative relationship between the MTB-1 ratio and AT in prior research
Table IV.
Likelihood of voluntary
SFAS 123 expense
recognition
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient (3) Coeffcient
Intercept ?5.095*** (37.74) ?3.672*** (16.77) ?5.116*** (23.49)
ISSUEp ? 2.237* (3.58) 1.966 (2.63) 2.782* (3.77)
AQUp ? ?9.300*** (9.92) ?9.378*** (9.92) ?8.100*** (6.63)
DEBTEQ ? ?0.010 (0.01) 0.036 (0.12) 0.112 (1.10)
INTCOV ? 0.134 (0.24) 0.050 (0.03) 0.169 (0.33)
BONUS ? 0.663* (3.39) 0.813** (4.84) 0.631 (2.53)
CEOOWN ? ?2.284 (1.55) ?1.521 (0.69) 0.430 (0.05)
INSTp ? 0.023 (0.00) 0.228 (0.21) 0.826 (1.96)
SIZE ? 0.298*** (22.21) 0.277*** (17.37) 0.321*** (19.23)
PROFITp ? 4.382* (3.37) 2.460 (0.91) 2.798 (1.03)
NI_PF ? ?33.534** (4.43) ?28.776* (3.22) ?6.444 (0.12)
NI_PFxINSTp ? 35.165* (3.66) 30.431* (2.75) 3.084 (0.02)
BSVOL ? ?2.570*** (15.23) ?2.544*** (12.87)
MTB-1 ? ?0.082* (3.77)
AT ? 0.068*** (7.66)
SSKEW ?
Number of ESO-expensing
frms 240 240 189
Number of ESO disclosure
frms 1,534 1,531 1,275
McFadden pseudo R-squared 18.6% 20.0% 15.7%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the
numbers in brackets belowthe coeffcients are Wald chi-squares; coeffcients are fromlogit regressions
of an indicator variable denoting whether, for fscal year 2004, the frm voluntarily expensed ESOs on
the fnancial statements under SFAS 123 over disclosure only; there were 155 ESOexpense-recognizing
frms for fscal year 2004 that were in the S&P 1500 index; the number of expense-recognizing frms is
less than the potential 2 years ? 155 ? 310 due to missing data in ExecuComp and Compustat; ten
industry dummy variables are not shown for ease of presentation; McFadden pseudo R-squared values
are calculated using a spreadsheet developed by Glynn (2003); all variables are winsorized at 1 and 99
per cent of their respective distributions and are described in Table I
139
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(Pae et al., 2005; Beaver and Ryan, 2005; Roychowdhury and Watts, 2007; Lawrence
et al., 2012). Specifcally, a cleanly conservative accounting treatment of voluntarily
expensing ESOs on the fnancial statements is related differently to the MTB-1 ratio and
conditional conservatism, supporting Givoly et al.’s (2007) suggestion that relying on
any single proxy to assess a frm’s overall degree of conservatism could lead to
erroneous inferences. Although, there is a negative relationship between ATand MTB-1
ratio in this study, this may not be the case in other situations. The goal of this study is
to show how an actual accounting choice is related to AT and MTB-1.
Finally, the coeffcient [?
13
from equation (1)] on variable BSVOL in Table IV
(Column 2) is negative with a value ? ?2.570 and statistically signifcant at the 1 per
cent level, indicating that frms were less likely to voluntarily expense ESOs if their
returns had greater volatility. This supports the prediction that frms were concerned
with howmuch and howvariable their ESOexpense would be, and, arguably, this is the
most important variable in explaining the decision to voluntarily expense ESOs[17].
Similar values and signifcance levels are obtained for BSVOL using the ATmeasure of
conservatism in the main model (Table IV, Column 3).
7. Additional analysis and robustness tests of the main results
7.1 AT and controlling for MTB-1 ratio
Pae et al. (2005) recommend controlling for the MTB-1 ratio when using the ATmeasure
to proxy for conservatism, and, more recently, Lawrence et al. (2012) also support the use
of controlling for the MTB-1 ratio. Table V(Column 1) presents fndings fromestimating
equation (1), for fscal years 2003 and 2004, using both the MTB-1 ratio and ATmeasure
in the same logistic regression. The coeffcient on MTB-1 is ?0.057 and the coeffcient
on AT is 0.067, statistically insignifcant and signifcant at the 1 per cent level,
respectively. Therefore, controlling for the MTB-1 supports the main quantitative and
qualitative results regarding conditional conservatism.
7.2 Third proxy for conservatism
Prior literature suggests using multiple proxies for conservatism, as there is not a
generally accepted measure for conservatism (Givoly et al., 2007; Beatty et al., 2008;
Hui et al., 2009). In Table V (Column 2) are the regression results of equation (1) with
the conservatism proxy SKEW. SKEW is measured as the difference between cash
fows (COMPUSTAT #308) and earnings (COMPUSTAT #172) defated by total
assets (COMPUSTAT #6) summed over the prior fve fscal years, similar to Beatty
et al. (2008). This measure is selected because arguably it contains items related to
both the MTB-1 ratio and conditional conservatism. Accounting earnings, which are
part of SKEW, are determined using mandatory accounting standards (the MTB-1
ratio) and include decisions made by the frm to include losses in a timelier manner
than gains (conditional conservatism)[18].
The coeffcient on SKEW is ?0.224 but statistically insignifcant at the 10 per cent
level. This fnding is consistent with the results in Table IV of the MTB-1 ratio and
conditional conservatism working in opposite directions for the likelihood of ESO
expense recognition, as SKEW is related to both proxies of conservatism.
7.3 Exclusion of fnancial frms
Many fnancial services frms simultaneously announced their intention to recognize
ESO expenses on the fnancial statements. This is evidenced from the large proportion
ARJ
27,2
140
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of frms from this industry in the expense recognition sample from both this and the
Aboody et al. (2004) study. Such an industry-wide decision to expense ESOs could refect
factors other than those included in the model. Financial institutions also differ fromother
frms in their debt to equity relationship, and Aboody et al. (2004) note that they likely have
lower stock-based compensation expense and may have a different investor base.
Table VI presents the pooledcross-sectional logistic regressionfndings fromestimating
equation (1), for fscal years 2003 and 2004, after eliminating fnancial service frms. The
coeffcient on MTB-1 is ?0.070 and still statistically signifcant at the 10 per cent level
(Column 1). The R
2
decreases from20.2 per cent to 15.5 per cent, likely due to the decrease in
Table V.
Likelihood of voluntary
SFAS 123 expense
recognition-alternative
proxies
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient
Intercept ?5.053*** (23.01) ?3.790*** (16.60)
ISSUEp ? 2.629*(3.31) 2.224* (3.24)
AQUp ? ?8.090*** (6.66) ?9.494*** (9.52)
DEBTEQ ? 0.134 (1.58) 0.032 (0.09)
INTCOV ? 0.075 (0.06) 0.016 (0.00)
BONUS ? 0.658* (2.75) 0.766** (4.06)
CEOOWN ? 0.487 (0.07) ?2.241 (1.39)
INSTp ? 0.837 (2.02) ?0.013 (0.00)
SIZE ? 0.341*** (20.66) 0.251*** (14.43)
PROFITp ? 2.256 (0.68) 4.623* (2.88)
NI_PF ? ?5.690 (0.10) ?49.565*** (7.76)
NI_PFxINSTp ? 3.324 (0.03) 59.598*** (7.92)
BSVOL ? ?2.596*** (13.17) ?2.409*** (12.84)
MTB-1 ? ?0.057 (1.57)
AT ? 0.067*** (7.52)
SSKEW ? ?0.224 (0.29)
Number of ESO-expensing frms 189 235
Number of ESO disclosure frms 1,273 1,508
McFadden pseudo R-squared 20.2% 20.8%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the numbers
inbrackets belowthe coeffcients are Waldchi-squares; coeffcients are fromlogit regressions of anindicator
variable denoting whether, for fscal year 2004, the frm voluntarily expensed ESOs on the fnancial
statements under SFAS 123 over disclosure only; there were 155 ESO expense-recognizing frms for fscal
year 2004that were inthe S&P1500index; the number of expense-recognizingfrms is less thanthe potential
2 years ?155 ?310 due to missing data in ExecuComp and Compustat; ten industry dummy variables are
not shown for ease of presentation; McFadden pseudo R-squared values are calculated using a spreadsheet
developed by Glynn (2003); all variables are winsorized at 1 and 99 per cent of their respective distributions
and are described in Table I
141
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
sample size. In addition, the results from Columns 2 and 3, using AT and SKEW as
conservatism proxies, also support the main fndings. In summary, the fnancial services
industry does not affect the main quantitative and qualitative results[19].
7.4 Firms that voluntarily announced expensing of ESOs
Determination of the frms that voluntarily announced plans to expense ESOs is
accomplished by collecting voluntary ESO (Statement of Financial Accounting
Standards, SFAS 123, 1995) expense recognition announcements from press
Table VI.
Likelihood of SFAS 123
expense recognition-
excluding fnancial
services frms
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient (3) Coeffcient
Intercept ?1.935* (3.57) ?2.700** (5.49) ?2.329** (4.67)
ISSUEp ? 0.824 (0.25) 0.636 (0.10) 0.880 (0.25)
AQUp ? ?6.102** (4.75) ?4.437 (2.22) ?6.156** (4.20)
DEBTEQ ? ?0.058 (0.22) 0.041 (0.11) ?0.043 (0.11)
INTCOV ? 0.088 (0.08) 0.284 (0.91) ?0.067 (0.03)
BONUS ? 0.524 (1.22) 0.148 (0.09) 0.572 (1.27)
CEOOWN ? ?3.489 (2.49) ?1.836 (0.66) ?4.337* (3.12)
INSTp ? ?2.284*** (12.53) ?2.389*** (10.32) ?2.335*** (12.06)
SIZE ? 0.252*** (11.31) 0.289*** (12.78) 0.229*** (8.81)
PROFITp ? 8.287*** (7.49) 9.424*** (8.36) 10.706*** (10.56)
NI_PF ? ?24.991* (3.02) ?15.207 (0.65) ?44.118*** (5.65)
NI_PFxINSTp ? 7.814 (0.25) ?3.689 (0.03) 33.734 (2.18)
BSVOL ? ?0.965 (1.83) ?0.930 (1.48) ?0.920 (1.45)
MTB-1 ? ?0.070* (2.78)
AT ? 0.067*** (5.62)
SSKEW ? 0.055 (0.01)
Number of ESO-expensing frms 143 121 140
Number of ESO disclosure frms 1,427 1,199 1,406
McFadden pseudo R-squared 15.5% 17.1% 15.7%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the
numbers in brackets below the coeffcients are Wald chi-squares; coeffcients are from logit
regressions of an indicator variable denoting whether, for fscal year 2004, the frm voluntarily
expensed ESOs on the fnancial statements under SFAS 123 over disclosure only; there were 155
ESO expense-recognizing frms for fscal year 2004 that were in the S&P 1500 index; the number of
expense-recognizing frms is less than the potential 2 years. ? 155 ? 310 due to missing data in
ExecuComp and Compustat; ten industry dummy variables are not shown for ease of presentation;
McFadden pseudo R-squared values are calculated using a spreadsheet developed by Glynn (2003);
all variables are winsorized at 1 and 99 per cent of their respective distributions and are described
in Table I
ARJ
27,2
142
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
releases in the LexisNexis news retrieval service using search terms “stock options”
and “expense”. In addition, a signifcant portion of these companies were obtained
from the ESO announcing frms listed in Ciesielski (2004). The search in LexisNexis
was conducted for announcements from January 2002 through December 31, 2003,
from which 146 frms were obtained. Of these frms, 40 were either not included in
Compustat or not in the S&P 1500 index. Aboody et al. (2004) identify 155 frms that
announced plans to recognize the expense in 2002 and early 2003. During 2003 and
2004, few frms announced their intention to expense ESOs.
In untabulated results, using frms that announced they would be expensing ESOs,
provided results consistent with using frms that actually expensed ESOs during the
SFAS 123 period. The MTB-1 ratio is still negatively related and the AT measure still
positively related to frms’ likelihood of expensing ESOs during the voluntary period of
SFAS 123 at statistically signifcant levels.
7.5 Beginning-of-year AT and SKEW
In the main model, the MTB-1 variable is the beginning-of-period value, but the
other two conservatism proxies, AT and SKEW, are the end-of-period values. Many
frms’ decision to expense ESOs could have been made based on their
beginning-of-period (ex-ante) degree of conservatism. Equation (1) is run again, but
replaced with the beginning-of-period AT and SKEW separately. In untabulated
results, AT is still positively related to EXPD and SKEW is still statistically
insignifcant. Therefore, the main fndings are the same when using
beginning-of-period conservatism proxies.
8. Summary and conclusions
This study examines the relationship between frms’ decisions to expense ESOs under
the voluntary period of SFAS 123 and their beginning-of-period MTB-1 ratio and
conditional conservatism. Even though expensing of ESOs is a conservative accounting
treatment of an economic activity, it is not obvious fromprior literature howthe decision
would be related to the MTB-1 ratio and conditional conservatism.
During the period when expensing ESOs was voluntary, SFAS 123, the MTB-1 ratio
was less likely associated with ESO expense recognition, but conditional
conservatismwas more likely associated with ESOexpense recognition. The MTB-1
ratio fndings are consistent with the incentive and asymmetric information
explanations of conservatism, reducing the need for additional voluntary
information. Regarding conditional conservatism, results are consistent with
Roychowdhury and Watts’ (2007) term of a “buffer” in net assets for recording
subsequent losses, and Beaver and Ryan’s (2005) analytical relationship that shows
the MTB-1 ratio preempting conditional conservatism. These fndings add to the
literature on the negative relationship between the MTB-1 ratio and conditional
conservatism and show that conservatism plays a role in the disclosure versus
recognition decision. Furthermore, the results support using more than one measure
in studies that involve accounting conservatism because each proxy of
conservatism can represent different forms and motives of conservatism.
This study makes a connection between conservatism and a frm’s willingness to
incur real costs from their fnancial reporting. Many frms lobbied against
expensing stock options, refecting the widely held perception that there were
143
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
substantial costs in recognizing ESOs within the fnancial statements. In addition,
this study shows that frms took into account the amount and variability of the ESO
expense because stock return variability is negatively related to ESO expense
recognition during the SFAS 123 period.
As with all studies related to accounting conservatism, there is the limitation of
suitable proxies for conservatism. Caution is advised to not interpret the results of this
study as a general relationship between the MTB-1 ratio and conditional conservatism
because this study is over a period of only two years; the frms are taken from the S&P
1500 index; and the relationship of the MTB-1 ratio and conditional conservatism is
specifc to ESO expense recognition.
Notes
1. Firms’ overall or general conservatismcan be described as the degree to which accounting
decisions lower income. By this description, the decision to expense ESOs is a
discretionary conservative choice because it clearly decreases income. Aboody et al.
(2004, p. 128) state “Recognition of SFAS 123 expense is a conservative accounting choice
consistent with that modeled in Hughes and Levine (2003) because it unambiguously
lowers net income relative to the disclosure-only alternative”. However, it is not evident
howMTB-1 and conditional conservatismwould be related to this decision because of the
multifaceted nature of the MTB-1 ratio. For example, it can also represent growth frms
(Fama and French, 1992).
2. Voluntarily expensing of ESOs is determined by “OE” in Compustat Mnemonic STKOF,
which represents frms that were early adopters of SFAS 123R.
3. Even non-academics like the Association of Investment Management and Research have
stated repeatedly over the years that stock options should be expensed and strongly
supported the 1994 proposal of the Financial Accounting Standards Board (FASB) to
require companies to expense stock options in the income statement. Nevertheless,
prominent business leaders, e.g. Andrew Grove of Intel and Harvey Golub, formerly of
American Express, argued against expensing of ESOs on the fnancial statements (Guay
et al., 2003).
4. Conservatism is defned as early as 1924 by Bliss (1924, p. 110). His defnition included
“anticipate no profts, but anticipate all losses”. However, FAS?B’s Conceptual Framework
for Financial Reporting (2010) does not include the principle of accounting conservatism
because it distorts the neutralityof fnancial information. Nevertheless, conservatismremains
prevalent in practice (Lawrence et al., 2012).
5. Even though the MTB ratio is related to unconditional conservatism, this study does not
imply this proxy of conservatism to mean unconditional conservatism.
6. Option values are highly dependent and positively related to a frm’s stock return
volatility (Merton, 1973; Hull, 2006, Chapters 15, 16, and 19). Blacconiere et al. (2011)
examine the voluntary disclosure of the reliability of stock option compensation during
the SFAS 123 period. They fnd stock return volatility to be positively related to this type
of voluntary disclosure.
7. It is possible that some frms will expense ESOs because they expect it to become mandatory
(Schrand, 2004). However, when frms that expensed during fscal year 2003 are used, this did
not change the main fndings of the study. Both proxies, MTB-1 and AT, have the same sign
ARJ
27,2
144
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
andare signifcant at the 1 and10 per cent level, respectively. SKEWis still insignifcant at the
10 per cent level. The sample periodis not extendedbeyondtwo years because a frm’s general
degree of conservatism is likely not constant for long periods.
8. For additional testing, frms that announced that they would be expensing voluntarily
ESOs are also examined (Section 7.4). Determination of the frms that voluntarily intend
to expense ESOs is accomplished by collecting voluntary ESO (SFAS 123) expense
recognition announcements from press releases on LexisNexis news retrieval service
using search terms “stock options” and “expense”. In addition, a signifcant portion of
voluntarily expensing companies was obtained from the ESO-expensing companies
listed in Ciesielski (2004).
9. The model of Aboody et al. (2004) is reproduced below, where the dependent variable
RECOGNIZEis 1 if a frmvoluntarily expensed ESOs and 0 otherwise. OSDIRSTK(outstanding
director stock) is not included because the data are not in the ExecuComp data set that was
available and also exclude %OPTTOP5 (per cent options held by top fve paid executives). The
latter is excluded because including it in any of the logistic regressions causes the validity of the
model tobe questionable andthe MLEmaynot exist. Still, the questionable results whenincluding
%OPTTOP5 did not change the main fndings of this paper. In the Aboody et al. model,
%OPTTOP5 is statistically insignifcant with a p-value ?0.472.
RECOGNIZEj ? ?0 ?1
N
INDNj ? ?1ISSUEj ? ?2ACQj ? ?3DEBT Eqj ??4INT COVj ?
?5BONUSj ? ?6CEO OWNj ??7OSDIRSTKj ? ?8INSTITj ? ?9SIZEj ??10PROFITj ?
?11%OPT TOP 5 j ??12COMPEXj ??13COMPEXj ?INSTITj ??j.
The primary fndings of Aboody et al. (2004) are that all their variables, except the last four, are
statistically signifcant at the 10 per cent level or better.
10. The ten industry groupings used as dummy variables are Mining/Construction/Extractive,
Food, Textiles/Print/Publish, Chemicals, Manufacturing, Computers, Transportation,
Utilities/Financial/Insurance and Retail. The Services industry is used as the control group.
The tables do not include the industry dummies for ease of presentation. The full table is
available on request from the author.
11. Guay et al. (2003, p. 408) note that empirical evidence shows frms that most actively lobbied
against ESO expensing are characterized by top executives having a greater portion of their
compensation from options. It is possible that CEOOWN will be negatively related to ESO
expense recognition.
12. For missing values in ExecuComp, the standard deviation of the prior fve years’ returns is
used. ExecuComp calculates BSVOL as the standard deviation calculated over 60 months.
13. As ?
2
in the denominator can be a very small number leading to extreme values for AT, AT
is restricted to values from zero to ten.
14. For ESO-expensing frms: CEOinformation is hand-collected fromproxy statement DEF 14A
for missing data in ExecuComp and INTCOV information is obtained from S&P’s Capital IQ
for missing data in Compustat. For the two years tested, there are 3000 S&P 1500 frms. Of
these, only 2,687 have an MTB ratio within Compustat. Given that approximately 10 per cent
of the frms voluntarily expensed ESOs in 2004, this would mean that we have 269 expensing
frms that have a MTBratio. However, there were still some frms for which CEOinformation
from the DEF 14A proxy statement was not available. These are the main reasons the fnal
sample for expensing frms when using the MTB ratio is only 240 frms.
145
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
15. Acomparison to the Aboody et al. (2004) would not be very meaningful because they use fscal
year 2001 for examination and some variable defnitions are different to the ones used this
study. For example, ISSUEp and AQUp are indicator variables in the Aboody et al. model but
are continuous variables in this study. Also, the Aboody et al. paper uses frms that
announced they would expense ESOs. The main fndings of this paper use frms that actually
expensed ESOs during the voluntary SFAS 123 period.
16. The logistic regression with AT produced a Quasi-complete separation when using only two
years of observations for the complete logistic regression. I therefore performed Firth’s
penalized likelihood approach, which is a method of addressing issues of separability, small
sample sizes and bias of the parameter estimates. Heinze and Schemper (2002) fnd this
approach an “ideal solution to separation” for logistic regression. In any case, the coeffcients
and statistical signifcance are very similar for AT whether the Firth approach is used or not
for the regression including AT.
17. The McFadden pseudo R
2
is the second highest, at 15.7 per cent, when only using the industry
dummy variables and BSVOL. The only variable by itself which produces a higher McFadden
pseudo R
2
, at 16.5 per cent, is SIZE.
18. Many studies, for example Beatty et al. (2008), use a composite measure for conservatism. A
composite measure is also testedinthis studyusingequal weights of AT, MTB-1 andSKEW. The
method is to frst convert each of these conservatism proxies to deciles so they are given equal
weight and assign one-third weight to each of the proxy deciles. A minimum of two proxies is
required and if only two are available for a specifc frm-year, then [one-half] weight is assigned.
The composite measure, when used in the logistic regression of equation (1) as the conservatism
proxy, also gave statistically insignifcant results (p-value ?0.67). This was expected because the
AT and MTB-1 conservatism proxies were oppositely related to the decision to expense ESOs.
This study calculates ATas (?
2,it
??
3,it
/?
2,it
). ATis also tested using only ?
3
fromequation (2),
which is also used in Givoly et al. (2007). The coeffcient on the decile ranking of ?
3
when used to
determine its relationship to the decision to voluntarily expense stock options is 0.047 with a two
sided p-value of 0.115.
19. I also test the model using a matched size sample of expensing and non-expensing frms.
Again the results support the main fndings.
References
Aboody, D., Barth, M. and Kasznik, R. (2004), “Firms voluntary recognition of stock-based
compensation expense”, Journal of Accounting Research, Vol. 42 No. 2, pp. 123-150.
Accounting Principles Board Opinion No. 25 (1972), Accounting for Stock Issued to Employees,
APB, New York, NY.
Ahmed, A. and Duellman, S. (2007), “Accounting conservatism and board of director
characteristics: an empirical analysis”, Journal of Accounting and Economics, Vol. 43
Nos 2/3, pp. 411-437.
Ball, R. and Shivakumar, L. (2005), “Earnings quality in UK private frms: comparative loss
recognition timeliness”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 83-128.
Barth, M.E., Gow, I.D. and Taylor, D.J. (2012), “Why do pro forma and street earnings not refect
changes in GAAP? Evidence fromSFAS 123R”, Reviewof Accounting Studies, Vol. 17 No. 3,
pp. 526-562.
Basu, S. (1997), “The conservatism principle and the asymmetric timeliness of earnings”, Journal
of Accounting and Economics, Vol. 24 No. 1, pp. 3-37.
ARJ
27,2
146
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Beatty, A. (2007), “Discussion of asymmetric timeliness of earnings, market-to-book and
conservatism in fnancial reporting”, Journal of Accounting and Economics, Vol. 44 No. 1,
pp. 32-35.
Beatty, A., Weber, J. and Yu, J.J. (2008), “Conservatism and debt”, Journal of Accounting and
Economics, Vol. 45 Nos 2/3, pp. 154-174.
Beaver, W. and Ryan, S. (2005), “Conditional and unconditional conservatism, concepts and
modeling”, Review of Accounting Studies, Vol. 10 Nos 2/3, pp. 269-309.
Blacconiere, G., Frederickson, J., Johnson, M. and Lewis, M. (2011), “Are voluntary disclosures
that disavow the reliability of mandated fair value information informative or
opportunistic?”, Journal of Accounting and Economics, Vol. 52 Nos 2/3, pp. 235-251.
Black, F. and Scholes, M. (1973), “The pricing of options and corporate liabilities”, Journal of
Political Economy, Vol. 81 No. 3, pp. 637-654.
Bliss, J.H. (1924), Management Though Accounts, The Ronald Press Company, New York, NY.
Bodie, Z., Kaplan, R. and Merton, R.C. (2003), “For the last time, stock options are an expense”,
Harvard Business Review, Vol. 81 No. 3, pp. 63-73.
Botosan, C. and Plumlee, M. (2001), “Stock option expense: the sword of Damocles revealed”,
Accounting Horizons, Vol. 15 No. 4, pp. 311-328.
Ciesielski, J. (2004), “S&P 500 companies electing to expense stock option compensation”, The
Analyst’s Accounting Observer Weblog, January, 12, 2004, Subscriber only Blog.
Dietrich, J.R., Muller, K. A. III and Riedl, E.J. (2007), “Asymmetric timeliness tests of accounting
conservatism”, Review of Accounting Studies, Vol. 12 No. 1, pp. 95-124.
Fama, E. and French, K. (1992), “The cross-section of expected stock returns”, Journal of Finance,
Vol. 47 No. 2, pp. 427-465.
Feltham, J. and Ohlson, J. (1995), “Valuation and clean surplus accounting for operating and
fnancial activities”, Contemporary Accounting Research, Vol. 11 No. 2, pp. 689-731.
Fields, T., Lys, L. and Vincent, L. (2001), “Empirical research on accounting choice”, Journal of
Accounting and Economics, Vol. 31 No. 1/3, pp. 255-307.
Francis, J., LaFond, R., Olsson, P. and Schipper, K. (2004), “Costs of equity and earnings
attributes”, The Accounting Review, Vol. 79 No. 4, pp. 967-1010.
Gigler, B. and Hemmer, T. (2001), “Conservatism, optimal disclosure policy, and the timeliness of
fnancial reports”, The Accounting Review, Vol. 76 No. 4, pp. 471-493.
Givoly, D., Hayn, C. and Natarajan, A. (2007), “Measuring reporting conservatism”, The
Accounting Review, Vol. 82 No. 1, pp. 65-106.
Glynn, P. (2003), “Spreadsheet to calculate McFadden pseudo R
2
”, available at:
http://staff.washington.edu/glynn/r2pseudo.xls (accessed 11 November 2012).
Guay, W., Kothari, S.P. and Sloan, R. (2003), “Accounting for employee stock options”, The
American Economic Review, Vol. 92 No. 2, pp. 405-409.
Heinze, G. and Schemper, M. (2002), “A solution to the problem of separation in logistic
regression”, Statistics in Medicine, Vol. 21 No. 16, pp. 2409-2419.
Hsu, A., O’Hanlon, J. and Peasnell, K. (2012), “The Basu measure as an indicator of conditional
conservatism: evidence from UK earnings components”, European Accounting Review,
Vol. 21 No. 1, pp. 87-113.
Hughes, J.S. and Levine, C.B. (2005), “Management compensation and earnings-based covenants
in resolving adverse selection in credit markets”, Journal of Applied Corporate Finance,
Vol. 11, pp. 832-850.
147
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hui, K.W., Matsunaga, S. and Morse, D. (2009), “The impact of conservatism on management
earnings forecasts”, Journal of Accounting and Economics, Vol. 47 No. 3, pp. 192-207.
Hull, J.C. (2006), Options, Futures and Other Derivatives, 6th ed., Prentice Hall, Upper Saddle River,
NJ.
Jenkins, D.S., Kane, G.D. and Velury, U. (2009), “Earnings conservatism and value relevance
across the business cycle”, Journal of Business Finance and Accounting, Vol. 39 Nos 9/10,
pp. 1041-1058.
Kinney, W. and Uecker, W. (1982), “Mitigating the consequences of anchoring in auditor
judgements”, The Accounting Review, Vol. 57 No. 1, pp. 55-69.
LaFond, R. and Watts, R. (2008), “The information role of conservatism”, The Accounting Review,
Vol. 83 No. 2, pp. 447-478.
Lawrence, A., Sloan, R. and Sun, Y. (2012), “Mandatorily conservative accounting, evidence and
implications”, working paper, U. C. Berkeley, Berkeley, CA.
Mason, L. (2004), “The impact of accounting conservatism on the magnitude of the differential
information content of cash fows and accruals”, Journal of Accounting, Auditing and
Finance, Vol. 19 No. 3, pp. 249-282.
Merton, R. (1973), “Theory of rational option pricing”, Bell Journal of Economics, Vol. 4 No. 1,
pp. 141-183.
Nasev, J. (2009), Conditional and Unconditional Conservatism Implications for Accounting Based
Valuation and Risky Projects, Gabler, Wiesbaden.
Pae, J., Thornton, D. and Welker, M. (2005), “The link between earnings conservatism and the
market-to-book ratio”, Contemporary Accounting Research, Vol. 22 No. 3, pp. 693-717.
Penman, S. and Zhang, X. (2002), “Accounting conservatism, the quality of earnings, and stock
returns”, The Accounting Review, Vol. 77 No. 2, pp. 237-264.
Qiang, X. (2007), “The effects of contracting, litigation, regulation, and tax costs on conditional
and unconditional conservatism: cross-sectional evidence at the frm level”, The
Accounting Review, Vol. 82 No. 3, pp. 759-796.
Roychowdhury, S. and Watts, R. (2007), “Asymmetric timeliness of earnings, market-to-book and
conservatismin fnancial reporting”, Journal of Accounting and Economics, Vol. 44 Nos 1/2,
pp. 2-31.
Rubinstein, M. (1995), “On the accounting valuation of employee stock options”, Journal of
Derivatives, Vol. 3 No. 1, pp. 8-24.
Ruddock, C., Taylor, S.J. and Taylor, S.L. (2006), “Nonaudit services and earnings conservatism: is
auditor independence impaired?”, Contemporary Accounting Research, Vol. 23 No. 3,
pp. 701-746.
Schrand, C. (2004), “Discussion of frms’ voluntary recognition of stock-based compensation
expense”, Journal of Accounting Research, Vol. 42 No. 2, pp. 151-158.
Statement of Financial Accounting Standards (SFAS) (1995), “Accounting for stock-based
compensation”, Summary of Statement No. 123, available at: www.fasb.org/st/
Watts, R.L. (2003a), “Conservatism in accounting part1: explanations and implications”,
Accounting Horizons, Vol. 17 No. 3, pp. 207-221.
Watts, R.L. (2003b), “Conservatism in accounting part II: evidence and research opportunities”,
Accounting Horizons, Vol. 17 No. 4, pp. 287-301.
Zhang, J. (2008), “The contracting benefts of accounting conservatismto lenders and borrowers”,
Journal of Accounting and Economics, Vol. 45 No. 1, pp. 27-54.
ARJ
27,2
148
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Further reading
Statement of Financial Accounting Standards Board (2004), “Share based payment”, Summary of
Statement No.123(R), available at: www.fasb.org/st/
About the author
Gulraze Wakil, BCOMM (U. of Toronto), MA (Economics-U. of Toronto), PhD, CGA received his
PhD in accounting (minor in fnance) from Kent State University, Ohio. He is an Assistant
Professor in the Sprott School of Business at Carleton University, Ottawa, Canada. His research
projects include accounting conservatism, accounting and capital markets, time-series forecasting
and international accounting. Teaching interests include all areas of accounting. He earned his
Certifed General Accountant Designation while working as an accountant in the insurance
industry in Toronto, Canada. Gulraze Wakil can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
149
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_823226927.pdf
This paper aims to examine the relationship between firms’ decisions to expense employee
stock options (ESOs) under the voluntary period of Statement of Financial Accounting Standard No. 123
(SFAS 123) and their market-to-book (MTB-1) ratio and conditional conservatism. Conservatism is
chosen because, even though expensing of ESOs is a conservative accounting treatment, it is not
obvious how the decision would be related to the MTB-1 ratio and conditional conservatism,
particularly because the MTB-1 is a long-run measure of conservatism and the decision to voluntarily
expense is examined over two years. The setting provides a unique opportunity to assess how two
major proxies of conservatism are related.
Accounting Research Journal
Market-to-book ratio and conditional conservatism: firms’ voluntary expensing of
employee stock options
Gulraze Wakil
Article information:
To cite this document:
Gulraze Wakil , (2014),"Market-to-book ratio and conditional conservatism: firms’ voluntary expensing of
employee stock options", Accounting Research J ournal, Vol. 27 Iss 2 pp. 124 - 149
Permanent link to this document:
http://dx.doi.org/10.1108/ARJ -11-2012-0088
Downloaded on: 24 January 2016, At: 21:20 (PT)
References: this document contains references to 48 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 520 times since 2014*
Users who downloaded this article also downloaded:
Prodosh Simlai, (2014),"Firm characteristics, distress risk and average stock returns", Accounting Research
J ournal, Vol. 27 Iss 2 pp. 101-123 http://dx.doi.org/10.1108/ARJ -06-2012-0046
Etumudon Ndidi Asien, (2014),"Impact of firm-specific characteristics on managers’ identity disclosure",
Accounting Research J ournal, Vol. 27 Iss 2 pp. 150-168 http://dx.doi.org/10.1108/ARJ -03-2013-0010
Ning Du, J ohn E. McEnroe, Kevin Stevens, (2014),"The joint effects of management incentive and
information precision on perceived reliability in fair value estimates", Accounting Research J ournal, Vol. 27
Iss 2 pp. 188-206 http://dx.doi.org/10.1108/ARJ -10-2012-0081
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Market-to-book ratio and
conditional conservatism: frms’
voluntary expensing of employee
stock options
Gulraze Wakil
Sprott School of Business, Carleton University, Ottawa, Canada
Abstract
Purpose – This paper aims to examine the relationship between frms’ decisions to expense employee
stock options (ESOs) under the voluntary period of Statement of Financial Accounting Standard No. 123
(SFAS 123) and their market-to-book (MTB-1) ratio and conditional conservatism. Conservatism is
chosen because, even though expensing of ESOs is a conservative accounting treatment, it is not
obvious how the decision would be related to the MTB-1 ratio and conditional conservatism,
particularly because the MTB-1 is a long-run measure of conservatism and the decision to voluntarily
expense is examined over two years. The setting provides a unique opportunity to assess how two
major proxies of conservatism are related.
Design/methodology/approach – Using frms listed in the S&P 1500 index, frms that expensed
ESOs are compared to frms that disclosed only in the fnancial statements. The main comparison uses
a logistic regression.
Findings – During the period when expensing ESOs was voluntary, SFAS 123, the MTB-1 ratio was
negatively associated with ESO expense recognition, but conditional conservatism was positively
associated with ESO expense recognition. The former is attributed to incentives of conservatism and
the latter to the tenet of conservatism tending to reduce income.
Research limitations/implications – The fndings add to the literature/controversy on the
negative relationship between the MTB-1 ratio and conditional conservatism. More important, they
support using more than one measure of conservatismin studies that involve accounting conservatism.
A limitation is that the fndings are specifc to voluntary ESO expense recognition.
Originality/value – This is the frst study to examine how conservatism affects the choice to
recognize an itemon the fnancial statements or disclosure only. In addition, the study shows that frms
were willing to incur real costs from their fnancial reporting.
Keywords Accounting conservatism, Employee stock options, SFAS 123
Paper type Research paper
1. Introduction
This paper investigates the relationship of two accounting conservatism measures, the
beginning-of-period market-to-book (MTB-1) ratio and conditional accounting
The author thanks Pervaiz Alam, Merridee Bujaki, Wei Wei Li, Michael R. Melton, Rafat Noor and
seminar participants at Kent State University. Also, the suggestions andcomments receivedat the
American Accounting Association NE region 2008 conference and Ohio region 2009 conference
are appreciated. This paper was awarded the best PhD student manuscript in 2009 by the AAA
Ohio region.
The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1030-9616.htm
ARJ
27,2
124
Accounting Research Journal
Vol. 27 No. 2, 2014
pp. 124-149
©Emerald Group Publishing Limited
1030-9616
DOI 10.1108/ARJ-11-2012-0088
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
conservatism with the choice of whether to recognize employee stock option (ESO)
expense on the fnancial statements or only disclosure in the notes. Conditional
conservatism is proxied by Basu’s (1997) asymmetric timeliness of earnings (AT)
measure. The choice was available due to Statement of Financial Accounting
Standard No. 123 (SFAS 123; Statement of Financial Accounting Standards, 1995),
which up until June 15, 2005, allowed companies to either recognize ESO expense or
to disclose it in the notes to the fnancial statements. This setting provides a unique
opportunity to disentangle how the MTB-1 ratio and conditional conservatism were
related to the conservative treatment of voluntarily expensing ESOs on the fnancial
statements[1].
Examining the MTB-1 ratio, which has been used as a measure of conservatism, and
conditional conservatism is of particular interest because of the controversy in the
literature over how they are related and the need for more research (Pae et al., 2005;
Beatty, 2007; Roychowdhury and Watts, 2007; Nasev, 2009). This study adds to the
literature by examining how the two proxies of conservatism are related to frms’
voluntary disclosure decisions on the fnancial statements; specifcally, the decision to
voluntarily expense ESOs in the fnancial records.
The empirical fndings of this study are consistent with predictions using both
actual and announcing voluntary ESO-expensing frms[2]. Regarding the MTB-1
ratio, frms which voluntarily expensed ESOs during the SFAS 123 period had lower
levels of the MTB-1 ratio. This result supports the incentive and asymmetric
information explanations of conservatism, reducing the need for additional
voluntary information. Regarding conditional conservatism, proxied by Basu’s
(1997) AT, frms which voluntarily expensed ESO during the SFAS 123 period had
higher degrees of conditional conservatism. These results are consistent with
Roychowdhury and Watts’ (2007) term of a “buffer” in net assets from recording
subsequent losses, Beaver and Ryan’s (2005) analytical relationship that shows the
MTB-1 ratio preempts conditional conservatism and Qiang (2007, p. 760) stating
that the MTB-1 ratio “immunizes” the accounting system from future bad news.
Also, conditional conservatism is more of a short-term measure of conservatism.
Further support of both results comes from Pae et al. (2005) and Givoly et al. (2007),
who also show a negative relationship between the MTB-1 ratio and subsequent
conditional conservatism, proxied by AT.
To summarize, this study contributes to the literature on accounting
conservatism and accounting choice in three ways. First, by using a specifc fexible
accounting standard, it shows that both the MTB-1 ratio and conditional
conservatismwere related to a frm’s decision to expense ESOs. Second, the fndings
of this study suggest that during short periods, the MTB-1 ratio and conditional
conservatism can be related in opposite directions, supporting Givoly et al.’s (2007)
views that the two types of conservatism capture different dimensions of
conservatism and that relying on only one measure of conservatism could lead to
erroneous conclusions. Third, this study makes a connection between conservatism
and a frm’s willingness to incur real economic costs from their fnancial reporting
behavior. There were costs to expensing ESOs because many frms lobbied against
expensing stock options, and Guay et al. (2003) discuss real economic costs to be a
possible reason for frms’ opposition.
125
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
The remainder of this paper is organized as follows: Section 2 provides background
on ESO accounting standards and presents defnitions and measures of conservatism.
Section 3 develops the empirical predictions, and Section 4 contains the sample selection
process. Section 5 presents the research model, and Section 6 provides the results and
fndings. In Section 7, additional analysis and robustness tests are presented, followed
by conclusions in Section 8.
2. Related prior literature
2.1 Evolution of standards on stock option accounting
The initial accounting standard for stock-based compensation was Accounting
Principles Board Opinion (APB) 25 (1972), Accounting for Stock Issued to Employees,
issued in 1972. This standard permitted companies to account for ESOs using the
“intrinsic value method”, which is the difference between the stock price and the option
exercise price at grant date, without subsequent changes in value. Given that most frms
grant options at an exercise price equal to the current stock price, the ESO expense is
zero using the intrinsic value method (Barth et al., 2012).
Eventually, the Financial Accounting Standards Board (FASB) issued exposure
draft #127-C on June 30, 1993, proposing the requirement of expensing of stock
options using either the Black–Scholes formula or binomial lattice methods for
option pricing at grant date and subsequent adjustments. Many hi-tech frms and
start-up businesses that relied heavily on stock options as an incentive in recruiting
and motivating employees, opposed the move. In contrast, most of the academic
literature agrees that ESOs should be expensed (Botosan and Plumlee, 2001; Bodie
et al., 2003; Guay et al., 2003). For example, Guay et al. (2003, p. 409) conclude:
“Accounting should refect the true costs of doing business, and labor acquired
through ESO grants is a real economic cost that frms should deduct from earnings
as an expense”. Similarly, Bodie et al. (2003) demonstrate that stock option grants
have real cash-fow implications that need to be reported, and detail distortions that
relegating stock option accounting to footnotes creates[3]. Nevertheless, until June
of 2005, frms were allowed to choose whether to recognize the expense or disclose it
in the footnotes of the fnancial statements (SFAS 123, issued October 1995). Almost
all companies decided on disclosure only because until the middle of 2002, just two
companies on the Fortune 500, Boeing and Winn Dixie, had decided to voluntarily
recognize ESO expense. From the middle of 2002 to the frst quarter of 2003, a
signifcant number of companies reported their intention to voluntarily recognize
ESO expense on their fnancial statements. Still, the majority of public companies
chose to only disclose the effects of ESOs in the notes to the fnancial statements
during 2002 through 2004.
SFAS 123(R) was released in December of 2004 and effective for fscal years
beginning June 15, 2005 (see Figure 2 in Section 3). This revision eliminated the option
companies had of only disclosing their ESO expense. From then, all companies, except
small business issuers, were required to use the fair-value method of calculating and
recognizing ESO expense on the fnancial statements as originally prescribed by
Statement 123. This revised standard was developed by the FASBto create a more level
playing feld in terms of management incentive compensation and its impact on a
company’s net income by requiring all forms of compensation to be expensed on the
fnancial statements.
ARJ
27,2
126
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.2 Valuation and recognition of ESOs
ESOs have many features that are different to exchange-traded options for which the
Black–Scholes formula was developed (Black and Scholes, 1973). For example, ESOs
usually have a vesting period during which they cannot be exercised, they have
non-transferability, their maturity is much longer (typically ten years) and they can be
exercised before an expiry date. Also, a common reason for giving ESOs is to align
employees’ and executives’ interests with those of the shareholders. Rubinstein
(1995) provides an extensive examination of the various methods and adjustments
needed for valuation of ESOs. He also critiques Exposure Draft 127-C (1993) and
provides additional methods for valuation, which he shows to be better than the
recommendations of the exposure draft.
2.3 Choice in accounting
Fields et al. (2001, p. 256) defne accounting choice as follows:
An accounting choice is any decision whose primary purpose is to infuence (either in form or
substance) the output of the accounting system in a particular way, including not only
fnancial statements published in accordance with GAAP, but also tax returns and regulatory
flings.
They conclude that there is considerable literature on accounting choice. However,
despite improvements in research designs, data sources and computing power, many
questions remain unanswered. The accounting choice dimension this paper investigates
is the voluntary decision to record ESO expense on the fnancial statements or disclose
in the notes to the fnancial statements.
A closely related paper to this study is the one by Hui et al. (2009). They fnd that
conservatism is negatively associated with the issuance of voluntary management
earnings forecasts. However, they do not differentiate between the MTB-1 ratio and
conditional conservatism. More specifc to this study is the paper by Aboody et al.
(2004), who investigate why frms would voluntarily choose to expense options. They
presume that frms would only make this choice if it were cost benefcial and, based on
this premise, identify factors that could explain the decision to expense (see Section 5 for
a discussion).
2.4 Conservatism
The offcial defnition of accounting conservatismis fromFASB Statement of Financial
Accounting Concepts (SFAC) No. 2 (1980)[4]. It describes conservatism as a prudent
reaction to uncertainty and to ensure that uncertainty and risks inherent in a business
situation are adequately considered. For example, if two estimates of amounts to be
received or paid in the future are about equally likely, conservatism requires the less
optimistic estimate to be used. Watts (2003a) and Basu (1997) describe conservatism as
the asymmetry in the verifcation requirements for gains and losses. Firms require a
greater degree of verifcation to record gains than to record losses. Another body of
literature explains conservatismas being induced by economic incentives. In particular,
Watts (2003a, p. 209) argues that conservatism is an “effcient technology employed in
the organization of the frm for its contracts with various parties”. The various parties
are external to the frm and have loss functions that are generally asymmetric. Watts
refers to this as the “contracting explanation” of conservatism and that conservative
accounting tends to reduce contracting costs. Numerous studies document the benefts
127
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
from conservatism in debt contracts (Beatty et al., 2008; Zhang, 2008) and equity
contracts (Ahmed and Duellman, 2007). In these articles, conservatism tends to reduce
income, which provides lenders and investors early signals of default risk, reducing
their contracting costs. As monitoring by external parties is costly, frms have an
incentive to commit to conservatism to get more favorable terms.
Beaver and Ryan (2005) defne unconditional (ex-ante or news-independent)
conservatism as the understatement of book value because of accounting concepts and
standards and conditional conservatism (ex-post or news-dependent) as the
discretionary writing down of assets under adverse conditions[5]. The current paper
interprets unconditional (conditional) conservatism as more of a long (short)-run stock
(fow) type of conservatism that is better represented in the balance sheet (income
statement). Examples of unconditional conservatism include accounting depreciation
that is greater than the actual depreciation of long-term assets, requirements for the
immediate expensing of certain items which last beyond one period and historical
cost accounting that prevents recording future benefts of positive net present value
projects. Jenkins et al. (2009, p. 1044) even state “Unconditional conservatism refers
to the reporting of conservative accounting numbers not conditioned on economic
reality”. Examples of conditional conservatism are how responsive the accounting
system is to new negative information regarding their existing receivables,
inventory and assets/liabilities in general. Figure 1 demonstrates how frms with
different levels of conditional conservatism would adjust their receivables to
information about their probability of collection. The more conditionally
conservative frm is represented by the curved line, which always reports net
receivables at a lower value, ceteris paribus.
2.5 Empirical conservatism measures
Watts (2003b) groups conservatism measures into three types:
(1) earnings and stock returns relation measures;
(2) net asset measures; and
(3) earnings and accrual measures.
100%
Pr. of Default
0 $ Net Receivables
Figure 1.
Conditional conservatism
and recording of
receivables
ARJ
27,2
128
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
2.5.1 Earnings and stock returns relation measures. Basu (1997) uses a regression of
earnings as a function of positive and negative returns as separate variables. He uses
positive stock returns to generally refect net asset gains and negative stock returns to
refect net asset losses. Basu argues that earnings will refect net losses more quickly
than net gains. This is the AT measure. The AT measure has been criticized as a
measure for conditional conservatism. For example, Givoly et al. (2007) fnd a
signifcantly negative association between the MTB-1 ratio and ATusing periods of 2.5,
5, and 10 years, and Dietrich et al. (2007) fnd biases in the coeffcient estimates of the AT
measure. However, Hsu et al.’s (2012) recent study supports it representing conditional
conservatism. Nevertheless, it is the most commonly used proxy for conditional
conservatism (Ball and Shivakumar, 2005; Beaver and Ryan, 2005; Pae et al., 2005;
Ruddock et al., 2006; Jenkins et al., 2009). Therefore, this study uses AT to proxy for
conditional conservatismand refers to it as a more short-termmeasure than the MTB-1
ratio, which is described in the second point below.
2.5.2 Net asset measures. The use of the MTB-1 ratio (market value divided by the net
book value of equity of the frm) as measure for conservatism emanates from the work
of Felthamand Ohlson (1995), whose defnition of conservatismis based on the disparity
between the market and book values of operating assets. Beaver and Ryan (2005, p. 269)
state, “We defne accountingconservatismas the onaverage understatement of the book
value of net assets relative to their market value (hereafter, the existence of expected
unrecorded goodwill)”. The MTB ratio is also used for conservatism in general. Nasev
(2009, p. 1) notes that the MTB ratio is the most obvious measure of conservatism.
Another study, by Pae et al. (2005), tries to reconcile conservatism in the income
statement with that in the balance sheet. They fnd conditional conservatism, the
tendency of frms to recognize bad news in earnings on a timelier basis than good news
(AT), to be negatively associated with the MTB-1 ratio (beginning-of-period MTBratio).
Roychowdhury and Watts (2007) explain how the correlation between the MTB-1 ratio
and AT can be negative, as observed by Pae et al. (2005), during short horizon periods.
They showthat both the MTB-1 ratio and ATmeasure conservatismwith error, and AT
is commonly not measured over horizons that include a frm’s initial public offering.
They demonstrate a positive association between the MTB-1 and AT over long time
horizons.
2.5.3 Accrual measures. Mason (2004) constructs a conservatismindex by assigning
values to the amount of R&D, advertising, inventory, unrecorded pension, accounts
receivable and depreciable assets. These are Mason’s proxies for the unrecorded
(accrual) value of a frm’s operating assets. Penman and Zhang (2002) also develop a
conservatismindex. They defne Ci ?ERi/NOAi, where Ci is the conservatismindex for
frmi, ER is the estimated reserve and NOAis the net operating assets. The ER is based
on the capitalization of R&D, advertising and differences in inventory between LIFO
and FIFO accounting policies. A complete conservatism index would include all
operating reserves, but the authors only choose these items because they are not subject
to management discretion.
3. Hypotheses development
The voluntary expensing of ESOs is consistent with the defnition(s) of conservatism
described in the literature review, and both the MTB-1 ratio and conditional
129
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
conservatism, proxied by AT, have been used to represent conservatism. Nevertheless,
the MTB-1 ratio and conditional conservatism represent different dimensions of
conservatism.
First, the MTB-1 ratio is the understatement of book values to their respective market
values, and a higher MTB-1 ratio implies greater past conservatism, ceteris paribus.
Given that expensing ESOs is a conservative accounting treatment, this supports a
positive relationship between expensing and the MTB-1 ratio. However, this positive
relationship is weakened because the MTB-1 ratio is a long-run measure of
conservatism that has been developing since the beginning of a company’s life. The
voluntary expensing of ESOs is only examined over a short period, 2003 and 2004, so a
frm’s short-run degree can be different from its long-run degree of conservatism (i.e.
MTB-1 ratio). This is even more likely for frms that have an increase in their market
value due to an increase in unrecorded intangibles that are unrelated to their degree
of conservatism. Further, frms that decided to voluntarily expense ESOs could have
had an incentive to do so because they have not been conservative in the past (i.e.
low MTB-1 ratio). This reasoning is supported by the contracting explanation of
conservatism whereby conservatism reduces information asymmetry to users
outside the frm (Watts, 2003a; LaFond and Watts, 2008). Furthermore, Gigler and
Hemmer (2001) develop theoretical links about the relationship between
conservatism and timely voluntary disclosures using a principal-agent model. They
fnd that less conservative frms would make more timely disclosures and thereby
provide greater voluntary information to the market. Similarly, Hui et al. (2009) fnd
a negative association between the MTB ratio and the issuance of voluntary
management earnings forecasts. They suggest that conservatism acts as a
substitute for management forecasts making conservative frms have less incentive
to voluntarily issue management forecasts.
Therefore, given the above arguments, the MTB-1 ratio’s relationship to the decision
to voluntarily expense ESOs is not obvious and therefore an empirical question. The
frst hypothesis for the MTB-1 ratio is presented in non-directional form:
H1. Ceteris paribus, during the period when expensing was not mandatory (SFAS
123), the beginning-of-period MTB-1 ratio is related to frms’ decision to
voluntarily expense stock options.
The second hypothesis is how conditional conservatism is related to the voluntary
expensing of ESOs. Conditional conservatism can be described as the recording of
income-decreasing economic events on a timelier basis than income-increasing
economic events. Clearly, the decision to expense ESOs is income-decreasing, which
supports a positive relationship. However, as discussed in formulating H1,
incentives also play a role in voluntary accounting choices. It was argued that frms
would have more incentive to expense if they had a low MTB-1 ratio. If this holds
true, then it adds more support for conditional conservatismbeing positively related
to expensing. Beaver and Ryan (2005) and Lawrence et al. (2012) fnd that
conditional conservatism is higher when frms have low MTB ratios. Therefore, a
positive relationship between the degree of a frm’s conditional accounting
conservatism, proxied by AT, and voluntary expensing of ESOs is predicted. The
second hypothesis is presented in directional form:
ARJ
27,2
130
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
H2. Ceteris paribus, during the period when expensing was not mandatory (SFAS
123), conditional accounting conservatismis positively related to frms’ decision
to voluntarily expense stock options.
Finally, although not a formal hypothesis, higher stock return volatility of a frm is
predicted to make a frm less likely to expense ESOs in favor of disclosure only on the
fnancial statements. This prediction is based on the assumption that frms would not
want to show a large amount for ESO expense, and large variations in comparative net
income may be interpreted as higher idiosyncratic risk for the frm[6]. In any case, not
including omitting this could lead to an omitted variable bias.
4. Data sources and periods
All available US frm-level fnancial data were obtained from Standard & Poor (S&P)’s
Research Insight Compustat database. However, the sample is restricted to frms in the
S&P 500, S&P 400 mid-capitalization and S&P 600 small-capitalization indices. This
group is selected because most of the ESO expense-recognizing frms are included in
these indices and covered by the ExecuComp database. Stock return and market
valuation data are obtained from the CRSP (Center for Research on Security Prices)
databases, and information regarding stock-based compensation and Black–Scholes
volatility is obtained from S&P’s ExecuComp database. For companies with missing
volatility numbers in ExecuComp, the volatility is calculated as the standard deviation
of annual returns over the past fve years. In addition, CEO data for ESO expensing
companies missing in ExecuComp are hand-collected from the Defnitive Notice &
Proxy Statement Schedule 14A.
The year selected for a frmto have voluntarily chosen to expense ESOs is fscal year
2004, fromwhich 155 (547) frms in the S&P 1500 index (Compustat population) had OE
in the Compustat Mnemonic STKCOF. OE in this Mnemonic signifes that the frmwas
an early adaptor of SFAS 123R. However, pooled cross-sectional testing is performed
over two fscal years – 2003 and 2004 – to allow for enough ESO-expensing frms to
obtain meaningful results from the conservatism proxies[7][8]. Therefore, the same
frms that expensed ESOs in fscal year 2004 are used for 2003. Figure 2 presents a
timeline that shows the various dates used for testing the hypotheses in this study and
for SFAS 123 and SFAS 123(R).
5. Research design
To test the hypotheses, frms that expensed ESOs are compared to frms that disclosed
only in the footnotes of the fnancial statements for the fscal year 2004. The main
comparison uses a logit regression of pooled cross-sectional frm-year observations for
fscal years 2003 and 2004. This is accomplished by using the same frms that expensed
ESOs in 2004 for years 2003 and 2004. The comparison group are frms in the S&P 1500
that did not expense ESOs.
The model used in this paper borrows heavily from the logit model of Aboody et al.
(2004), which examines likely factors associated with frms that voluntarily announced
Beg. fiscal yr. 2003 Beg. fiscal yr. 2004 End fiscal yr. 2004 June 15, 2005
Inclusion Year SFAS 123R
Hypotheses testing period required
Figure 2.
Timeline of SFAS 123 and
testing period
131
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that they would expense ESOs[9]. They categorize their independent variables into four
areas that are likely associated with expensing ESOs: capital market activity, private
incentives of management and the board of directors, political costs of not expensing
ESOs and controlling for the magnitude of SFAS 123 expense. The current paper
incorporates similar variables, including their predicted direction, in the model of this
study but also includes the MTB-1 ratio, conditional conservatism(AT) and stock return
volatility.
The model is presented below as equation (1):
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB ? 1, AT, SKEW
?
it ? ?
it
(1)
The dependent variable, EXPD, is a dichotomous variable that equals 1 for an ESO
expense-recognizing frm (Compustat Mnemonic STKCOF ? OE) and 0 for
disclosure-only frms for fscal year 2004. Next, IND
N
are indicator variables that equal
1 if the frmis in industry N, and 0 otherwise, based on SIC code classifcations grouped
into ten categories. The industry variables are to control for the potential effects of
omitted variables associated with industry membership[10].
The frst set of variables relate to capital market activity, which are predicted to be
positively related to SFAS 123 ESO expense recognition. ISSUEp is the percentage
change in common shares outstanding (Compustat #A25); AQUp is the ratio of cash
outfow of funds used for and/or the costs relating to acquisition of a company
(Compustat #A129) defated by total assets (COMPUSTAT #6); DEBTEQ
(Compustat Mnemonic DCE) is the ratio of long-term debt to common equity; and
INTCOV is the ratio of interest expense (Compustat #A15) to net operating income
(Compustat #A178).
The second set relates to private incentives of management. BONUS is the ratio of
CEO cash bonus to total cash compensation and predicted to be negatively related to
ESO expense recognition; CEOOWN is equity shares and options held by the CEO as a
percentage of total shares outstanding and predicted to be positively related to ESO
expense recognition[11]. In between the above set of variables and the ones in the next
paragraph is the percentage of shares outstanding held by institutional investors
(INSTp) used as a proxy for information asymmetry. It is predicted that the greater the
institutional investor percentage, the less likely the frm will expense ESOs because
institutional investors are more sophisticated, which enables themto better understand
footnote disclosure.
The next sets of variables are characterized as political costs of not expensing
ESOs and are predicted to have a positive relation with voluntary ESO expense
recognition. SIZE is the logarithm of market value of equity (Compustat Mnemonic
MKVALF); PROFITp is net income (Compustat #A172) defated by market value of
equity.
Consistent with the Aboody et al. (2004) model, the magnitude of SFAS 123 expense
is controlled by including NI_PF, which is the difference between reported net income
ARJ
27,2
132
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
and S&P core earnings (Compustat Mnemonic COREEARN), defated by market value
of equity. Finally, the interaction of NI_PF and INSTp is also included to control for
information asymmetry.
BSVOL is the Black–Scholes volatility measure obtained from the ExecuComp
database[12]. This variable is important in the expense recognition decision because a
highly volatile stock would increase the value and variability of ESOs, which may
discourage a company from expensing ESOs if they are concerned that this would
signifcantly affect their net income and its variability. Therefore, this variable is
predicted to be negatively related to ESO expense recognition. The fnal variables are
the conservatism proxies, which will be substituted individually by the
beginning-of-period MTB-1 ratio (Compustat Mnemonic MKBKF), AT measure for
conditional conservatism and skewness of cash fows to earnings (SKEW). The MTB-1
ratio attempts to capture conservatism that has been developing since the inception of
the frm. AT is calculated using the procedure described in the next paragraph and
SKEWis another proxy for conservatismwhich is discussed in Section 7.2 as additional
analysis.
Basu (1997) measured a frm’s degree of conservatismas the sensitivity of earnings to
negative returns. In this paper, Basu’s model of earnings as a function of
contemporaneous stock returns [equation (2)] is used for the measure of conditional
conservatism (AT).
X
it
/P
it?1
? ? ? ?
1
DR
it
? ?
2
R
it
? ?
3
(R
it
xDR
it
) ? ?
it
(2)
Where:
X
it
?Earnings per share before extra-ordinary items and discontinued operations
[Compustat item #58];
R
it
?annual cum dividend return from CRSP;
DR
it
? a dummy variable which equals 1, when returns are negative, and 0
otherwise;
P
it-1
?end of prior fscal year stock price [Compustat item #199].
The main coeffcient of interest is ?
3
, which represents the incremental effect on
earnings from negative returns (coined “bad news” in Basu’s article), in comparison to
positive returns (good news). Apositive and statistically signifcant ?
3
is interpreted as
existence of accounting conservatism because negative returns have a greater effect on
earnings than do positive returns. Following Francis et al. (2004) and Zhang (2008), the
sensitivity of earnings to negative returns (?
2
??
3
), relative to positive returns (?
2
), is
used as the proxy for frm-level conditional conservatism – AT ?(?
2,it
??
3,it
/?
2,it
)[13].
This quotient is calculated on a rolling ten-year window, t-9 […], t. A minimum fve
years of earnings and returns is required for a frm-year observation to be given an AT
measure of conservatism. It is expected that AT will be greater than one.
6. Empirical results
6.1 Descriptive statistics
Table I, panels Aand B, report the breakdown of the variables used in this paper. Panel
Apresents descriptive statistics for the whole sample and panel B has frms partitioned
by recognizing ESOs on the fnancial statements and disclosure-only frms. All
133
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table I.
Descriptive statistics for
conservatism proxies,
return volatility and
control variables
P
a
n
e
l
A
S
a
m
p
l
e
p
e
r
i
o
d
:
2
0
0
3
-
2
0
0
4
V
a
r
i
a
b
l
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
M
e
d
i
a
n
2
5
t
h
p
e
r
c
e
n
t
i
l
e
7
5
t
h
p
e
r
c
e
n
t
i
l
e
F
Y
R
1
,
7
7
4
2
,
0
0
3
.
5
0
.
5
0
0
2
,
0
0
4
.
0
2
,
0
0
3
.
0
2
,
0
0
4
.
0
I
S
S
U
E
p
1
,
7
7
4
0
.
0
2
2
0
.
0
6
4
0
.
0
1
0
0
.
0
0
0
0
.
0
2
6
A
Q
U
p
1
,
7
7
4
0
.
0
2
4
0
.
0
5
5
0
.
0
0
0
0
.
0
0
0
0
.
0
1
8
D
E
B
T
E
Q
1
,
7
7
4
0
.
6
2
0
0
.
8
7
2
0
.
4
0
4
0
.
1
0
1
0
.
8
0
5
I
N
T
C
O
V
1
,
7
7
4
0
.
1
9
6
0
.
3
8
1
0
.
1
0
9
0
.
0
3
0
0
.
2
6
3
B
O
N
U
S
1
,
7
7
4
0
.
4
4
3
0
.
2
4
4
0
.
4
9
5
0
.
3
1
2
0
.
6
1
9
C
E
O
O
W
N
1
,
7
7
4
0
.
0
3
5
0
.
0
5
6
0
.
0
1
6
0
.
0
0
8
0
.
0
3
5
I
N
S
T
p
1
,
7
7
4
0
.
8
6
5
0
.
1
6
0
0
.
8
8
7
0
.
7
7
2
0
.
9
6
9
S
I
Z
E
1
,
7
7
4
8
,
2
0
4
1
9
,
0
7
6
2
,
0
1
0
8
4
8
6
,
0
3
9
P
R
O
F
I
T
p
1
,
7
7
4
0
.
0
4
0
0
.
0
5
6
0
.
0
4
7
0
.
0
3
1
0
.
0
6
2
N
I
_
P
F
1
,
7
7
4
0
.
0
0
4
0
.
0
2
6
0
.
0
0
4
0
.
0
0
1
0
.
0
0
9
B
S
V
O
L
1
,
7
7
4
0
.
4
3
7
0
.
2
0
3
0
.
3
8
3
0
.
2
9
8
0
.
5
3
4
M
T
B
-
1
1
,
7
7
1
2
.
7
2
3
2
.
3
2
4
2
.
0
9
8
1
.
4
7
8
3
.
2
6
8
A
T
1
,
4
6
4
2
.
1
5
1
3
.
2
4
4
0
.
4
0
6
0
.
0
0
0
2
.
8
4
3
S
S
K
E
W
1
,
7
4
3
0
.
2
8
5
0
.
2
4
4
0
.
2
5
6
0
.
1
4
7
0
.
3
8
0
P
a
n
e
l
B
V
a
r
i
a
b
l
e
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
D
i
s
c
l
o
s
u
r
e
-
o
n
l
y
f
r
m
s
D
i
f
f
e
r
e
n
c
e
i
n
m
e
a
n
s
t
-
s
t
a
t
p
-
v
a
l
u
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
I
S
S
U
E
p
2
4
0
0
.
0
2
8
0
.
0
7
0
1
,
5
3
4
0
.
0
2
1
0
.
0
6
2
0
.
0
0
7
1
.
5
6
0
0
.
1
1
9
A
Q
U
p
2
4
0
0
.
0
0
8
0
.
0
2
7
1
5
3
4
0
.
0
2
7
0
.
0
5
8
?
0
.
0
1
8
?
4
.
8
3
0
?
.
0
0
0
1
D
E
B
T
E
Q
2
4
0
0
.
8
2
5
0
.
9
4
0
1
5
3
4
0
.
5
8
8
0
.
8
5
6
0
.
2
3
7
3
.
9
3
0
?
.
0
0
0
1
I
N
T
C
O
V
2
4
0
0
.
2
5
2
0
.
3
4
5
1
5
3
4
0
.
1
8
8
0
.
3
8
6
0
.
0
6
4
2
.
4
4
0
0
.
0
1
5
B
O
N
U
S
2
4
0
0
.
5
2
0
0
.
2
5
2
1
5
3
4
0
.
4
3
1
0
.
2
4
0
0
.
0
8
9
5
.
3
0
0
?
.
0
0
0
1
C
E
O
O
W
N
2
4
0
0
.
0
2
4
0
.
0
4
5
1
5
3
4
0
.
0
3
7
0
.
0
5
7
?
0
.
0
1
3
?
3
.
4
5
0
?
.
0
0
0
1
I
N
S
T
p
2
4
0
0
.
8
2
9
0
.
1
7
5
1
5
3
4
0
.
8
7
0
0
.
1
5
6
?
0
.
0
4
1
?
3
.
7
1
0
?
.
0
0
0
1
(
c
o
n
t
i
n
u
e
d
)
ARJ
27,2
134
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table I.
P
a
n
e
l
B
V
a
r
i
a
b
l
e
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
D
i
s
c
l
o
s
u
r
e
-
o
n
l
y
f
r
m
s
D
i
f
f
e
r
e
n
c
e
i
n
m
e
a
n
s
t
-
s
t
a
t
p
-
v
a
l
u
e
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
O
b
s
e
r
v
a
t
i
o
n
M
e
a
n
S
t
a
n
d
a
r
d
d
e
v
i
a
t
i
o
n
S
I
Z
E
2
4
0
1
6
,
7
9
6
3
1
,
1
9
7
1
5
3
4
6
,
8
6
0
1
5
,
9
9
3
9
,
9
3
6
7
.
6
2
0
?
.
0
0
0
1
P
R
O
F
I
T
p
2
4
0
0
.
0
5
4
0
.
0
4
7
1
5
3
4
0
.
0
3
7
0
.
0
5
7
0
.
0
1
6
4
.
2
9
0
?
.
0
0
0
1
N
I
_
P
F
2
4
0
0
.
0
0
4
0
.
0
2
9
1
5
3
4
0
.
0
0
4
0
.
0
2
5
0
.
0
0
0
0
.
1
4
0
0
.
8
8
5
B
S
V
O
L
2
4
0
0
.
3
2
5
0
.
1
5
7
1
5
3
4
0
.
4
5
5
0
.
2
0
3
?
0
.
1
3
0
?
9
.
4
5
0
?
.
0
0
0
1
M
T
B
-
1
2
4
0
2
.
3
6
4
2
.
0
2
4
1
5
3
1
2
.
7
7
9
2
.
3
6
3
?
0
.
4
1
5
2
.
5
7
0
0
.
0
1
0
A
T
1
8
9
2
.
6
9
1
3
.
5
6
9
1
2
7
5
2
.
0
7
1
3
.
1
8
7
0
.
6
2
0
2
.
4
6
0
0
.
0
1
4
S
S
K
E
W
2
3
5
0
.
2
1
8
0
.
2
0
3
1
5
0
8
0
.
2
9
5
0
.
2
4
8
?
0
.
0
7
7
?
4
.
5
5
0
?
.
0
0
0
1
N
o
t
e
s
:
I
S
S
U
E
p
i
s
t
h
e
p
e
r
c
e
n
t
a
g
e
c
h
a
n
g
e
i
n
c
o
m
m
o
n
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
(
#
A
2
5
)
;
A
Q
U
p
i
s
t
h
e
r
a
t
i
o
o
f
c
a
s
h
o
u
t
f
o
w
o
f
f
u
n
d
s
u
s
e
d
f
o
r
a
n
d
/
o
r
c
o
s
t
s
r
e
l
a
t
i
n
g
t
o
a
c
q
u
i
s
i
t
i
o
n
o
f
a
c
o
m
p
a
n
y
(
#
A
1
2
9
)
d
e
f
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
(
#
6
)
;
D
E
B
T
E
Q
(
M
n
e
m
o
n
i
c
D
C
E
)
i
s
t
h
e
r
a
t
i
o
o
f
l
o
n
g
-
t
e
r
m
d
e
b
t
t
o
c
o
m
m
o
n
e
q
u
i
t
y
;
I
N
T
C
O
V
i
s
t
h
e
r
a
t
i
o
o
f
i
n
t
e
r
e
s
t
e
x
p
e
n
s
e
(
#
A
1
5
)
t
o
n
e
t
o
p
e
r
a
t
i
n
g
i
n
c
o
m
e
(
#
A
1
7
8
)
;
B
O
N
U
S
i
s
t
h
e
r
a
t
i
o
o
f
C
E
O
c
a
s
h
b
o
n
u
s
t
o
t
o
t
a
l
c
a
s
h
c
o
m
p
e
n
s
a
t
i
o
n
;
C
E
O
O
W
N
i
s
e
q
u
i
t
y
s
h
a
r
e
s
a
n
d
o
p
t
i
o
n
s
h
e
l
d
b
y
t
h
e
C
E
O
a
s
a
p
e
r
c
e
n
t
a
g
e
o
f
t
o
t
a
l
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
;
I
N
S
T
p
i
s
t
h
e
p
e
r
c
e
n
t
a
g
e
o
f
s
h
a
r
e
s
o
u
t
s
t
a
n
d
i
n
g
h
e
l
d
b
y
i
n
s
t
i
t
u
t
i
o
n
a
l
i
n
v
e
s
t
o
r
s
;
S
I
Z
E
i
s
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
(
M
n
e
m
o
n
i
c
M
K
V
A
L
F
)
;
P
R
O
F
I
T
p
i
s
n
e
t
i
n
c
o
m
e
(
#
A
1
7
2
)
d
e
f
a
t
e
d
b
y
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
;
N
I
_
P
F
i
s
t
h
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
r
e
p
o
r
t
e
d
n
e
t
i
n
c
o
m
e
a
n
d
S
&
P
c
o
r
e
e
a
r
n
i
n
g
s
(
M
n
e
m
o
n
i
c
C
O
R
E
E
A
R
N
)
d
e
f
a
t
e
d
b
y
m
a
r
k
e
t
v
a
l
u
e
o
f
e
q
u
i
t
y
;
B
S
V
O
L
i
s
t
h
e
B
l
a
c
k
–
S
c
h
o
l
e
s
v
o
l
a
t
i
l
i
t
y
m
e
a
s
u
r
e
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
E
x
e
c
u
C
o
m
p
d
a
t
a
b
a
s
e
;
M
T
B
-
1
i
s
t
h
e
b
e
g
i
n
n
i
n
g
-
o
f
-
p
e
r
i
o
d
m
a
r
k
e
t
-
t
o
-
b
o
o
k
r
a
t
i
o
(
M
n
e
m
o
n
i
c
M
K
B
K
F
)
;
A
T
i
s
t
h
e
s
e
n
s
i
t
i
v
i
t
y
o
f
e
a
r
n
i
n
g
s
t
o
n
e
g
a
t
i
v
e
r
e
t
u
r
n
s
c
a
l
c
u
l
a
t
e
d
o
v
e
r
1
0
y
e
a
r
s
(
m
i
n
i
m
u
m
o
f
5
y
e
a
r
s
)
;
a
n
d
S
K
E
W
i
s
t
h
e
d
i
f
f
e
r
e
n
c
e
b
e
t
w
e
e
n
c
a
s
h
f
o
w
s
(
#
3
0
8
)
a
n
d
n
e
t
i
n
c
o
m
e
(
#
1
7
2
)
d
e
f
a
t
e
d
b
y
t
o
t
a
l
a
s
s
e
t
s
(
#
6
)
s
u
m
m
e
d
o
v
e
r
t
h
e
p
r
i
o
r
f
v
e
f
s
c
a
l
y
e
a
r
s
;
R
e
c
o
g
n
i
z
i
n
g
f
r
m
s
o
f
2
4
0
a
r
e
a
f
t
e
r
r
e
q
u
i
r
i
n
g
a
l
l
v
a
r
i
a
b
l
e
s
t
o
h
a
v
e
a
v
a
l
u
e
,
o
t
h
e
r
t
h
a
n
t
h
e
c
o
n
s
e
r
v
a
t
i
s
m
p
r
o
x
i
e
s
;
L
o
g
a
r
i
t
h
m
o
f
S
I
Z
E
i
s
u
s
e
d
i
n
r
e
g
r
e
s
s
i
o
n
s
.
A
l
l
c
o
n
t
i
n
u
o
u
s
v
a
r
i
a
b
l
e
s
a
r
e
w
i
n
s
o
r
i
z
e
d
a
t
1
a
n
d
9
9
p
e
r
c
e
n
t
o
f
t
h
e
i
r
r
e
s
p
e
c
t
i
v
e
d
i
s
t
r
i
b
u
t
i
o
n
s
135
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
continuous independent variables are winsorized at the top and bottom 1 per cent of
their respective distributions to mitigate the infuence of outliers.
The two main conservatism proxies examined in this study are the
beginning-of-period MTB-1 ratio and the AT, the latter representing conditional
conservatism. Panel Bshows the MTB-1 ratio of expensing compared to disclosure-only
frms is lower by 0.415 (18 per cent), and the AT measure of expensing compared to
disclosure-only frms is higher by 0.620 (23 per cent). Also shown in Panel B is the
SKEW measure of conservatism. It is lower for the ESO expense-recognizing frms and
is further discussed in Section 7.2. With respect to the control variables, differences
include a lower ACUp, higher DEBTEQ, higher INTCOV, higher BONUS, lower
CEOOWN, higher SIZE and higher PROFITp for the ESO-expensing frms. ISSUEp,
INSTp and NI_PF are similar between the two groups[14].
This study also examines how stock volatility was related to voluntarily expensing
ESOs. This variable, BSVOL, is 0.130 (29 per cent) lower for expensing frms, suggesting
that higher stock return volatility reduced the likelihood of ESO expense recognition.
Table II presents the industry classifcation for recognizing and disclosure-only
frms. A major difference between recognizing and disclosure-only frms is in the
fnancial services industry. Financial services constitute approximately 40 per cent of
Table II.
Industry classifcation
and descriptive statistics
Sample period: 2003-2004 Recognizing frms Disclosure-only frms
Industry Number (%) Number (%)
Other 2 0.83 11 0.72
1 Chemicals 9 3.75 57 3.72
2 Computers 3 1.25 193 12.58
3 Extractive 14 5.83 66 4.30
4 Financial 50 20.83 42 2.74
5 Food 8 3.33 86 5.61
6 Insurance/Real Estate 47 19.58 23 1.50
7 Manf: Electrical equipment 2 0.83 51 3.32
8 Manf: Instruments 0 0.00 108 7.04
9 Manf: Machinery 3 1.25 77 5.02
10 Manf: Metal 6 2.50 52 3.39
11 Manf: Misc. 0 0.00 19 1.24
12 Manf: Rubber/glass/etc. 2 0.83 27 1.76
13 Manf: Transport equipment 6 2.50 42 2.74
14 Mining/Construction 8 3.33 38 2.48
15 Pharmaceuticals 2 0.83 53 3.46
16 Retail: Misc. 15 6.25 111 7.24
17 Retail: Restaurant 2 0.83 29 1.89
18 Retail: Wholesale 10 4.17 59 3.85
19 Services 11 4.58 122 7.95
20 Textiles/Print/Publ 6 2.50 87 5.67
21 Transportation 10 4.17 66 4.30
22 Utilities 24 10.00 115 7.50
Totals 2 40 1 00
Notes: The above industries are summarized fromfour-digit SIC codes into 22 industries; Category 4
and 6, in italics, are considered fnancial services companies
ARJ
27,2
136
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
the recognizing frms but only 4 per cent of the disclosure-only frms. Additional testing
that excludes fnancial services frms is performed in Section 7.3.
6.2 Univariate analysis of the association between variables
Table III presents the Pearson and Spearman correlation coeffcients among all the
variables used in the model. The univariate relationships of whether a frm expensed
ESOs on the fnancial statements (EXPD) with the independent variables are generally
in the same direction as the multivariate results reported in the next section. Of more
interest to this study is how EXPD is related to the conservatism proxies. EXPD is
negatively related to the MTB-1 ratio, but positively related to the conditional
conservatism proxy (AT) with correlation coeffcients (?) of ?0.088 and 0.064,
respectively. In addition, correlation coeffcient between the MTB-1 ratio and AT is
negative with a (?) of ?0.035 with a p value of 0.178, suggesting that they are negatively
related for the current sample. Nevertheless, it is diffcult to draw defnitive
implications, as the univariate analysis ignores the potential impact of correlations
between other explanatory variables. Multivariate analysis is required to address the
main questions of this paper, which is mentioned in the next sub-section.
6.3 Analysis of logistic regression results
The results of the logistic model [equation (1)] for fscal years 2003 and 2004 are shown
in Table IV (Columns 1, 2 and 3). Column 1 presents the coeffcients (Wald chi-square)
for the variables used from the Aboody et al. (2004) model without the conservatism
proxies[15]. The main coeffcients of interest of this paper are fromvariables MTB-1 and
AT, in Table IV, Columns 2 and 3, respectively. First, the coeffcient on MTB-1 is
negatively related to whether a frm expensed ESOs with a value of ?0.082 and
statistically signifcant at the 10 per cent level. The hypothesis of the relationship
between EXPD and the MTB-1 ratio, proxied by MTB-1, was presented as
non-directional because the income-decreasing tendency of conservatismand incentives
to expense ESOs were in opposite directions. In any case, H1 is supported because the
MTB-1 ratio is related to EXPDat a statistically signifcant level. Ahigher MTB-1 ratio
is interpreted in this paper as being relatively more conservative in the past. Given that
the coeffcient on MTB-1 is negative suggests that frms which had higher MTB-1 ratios
had less of an incentive to expense ESOs. Conservatism is documented to reduce
information asymmetry, so it is possible that expensing ESOs would not materially help
to reduce information asymmetry for high MTB-1 ratio frms.
Second, the coeffcient on AT is positively related to a frm expensing ESOs with a
value of 0.068 and statistically signifcant at the 1 per cent level (Column 3), indicating
that voluntary expensing of ESOs was more likely associated with frms that were
conditionally conservative, supporting H2[16]. The arguments for conditional
conservatism being positively related to the likelihood of ESO expense recognition
rested mainly on conditional conservatism’s asymmetric requirements to record
income-reducing items in a timelier manner than income-increasing items. Expensing
ESOs will lower income compared to disclosure only. In addition, the hypothesis
development for H2 in Section 3 did not predict conditional conservatismto be related to
incentives for expensing ESOs.
The above two results show that the MTB-1 ratio is negatively related and AT
positively related to the voluntary expensing of ESOs. Apossible explanation for this is
137
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Table III.
Correlation analysis
among dependent and
independent variables
Pearson
(above)/Spearman (below)
V
a
r
i
a
b
l
e
E
X
P
D
I
S
S
U
E
p
A
Q
U
p
D
E
B
T
E
Q
I
N
T
C
O
V
B
O
N
U
S
C
E
O
O
W
N
I
N
S
T
p
S
I
Z
E
P
R
O
F
I
T
p
N
I
_
P
F
B
S
V
O
L
M
T
B
-
1
A
T
S
K
E
W
E
X
P
D
0
.
0
3
7
?
0
.
1
1
4
0
.
1
0
1
?
0
.
0
6
1
0
.
1
2
5
?
0
.
0
8
2
0
.
0
9
3
0
.
1
7
8
0
.
0
0
3
0
.
0
5
8
?
0
.
2
1
9
?
0
.
0
8
8
0
.
0
6
4
?
0
.
1
0
8
I
S
S
U
E
p
?
0
.
0
0
8
0
.
1
6
0
?
0
.
1
1
6
?
0
.
1
1
1
0
.
0
0
1
?
0
.
0
4
1
0
.
0
6
7
?
0
.
0
9
9
0
.
0
0
3
0
.
1
0
4
0
.
0
7
4
0
.
0
3
8
?
0
.
0
0
5
0
.
0
3
3
A
Q
U
p
?
0
.
1
2
3
0
.
0
2
6
?
0
.
0
0
4
0
.
0
3
3
0
.
0
2
5
0
.
0
3
5
0
.
0
0
1
?
0
.
0
6
4
0
.
0
0
0
?
0
.
0
8
8
0
.
0
1
7
0
.
0
9
5
?
0
.
0
3
9
0
.
0
3
5
D
E
B
T
E
Q
0
.
1
5
1
?
0
.
1
1
6
?
0
.
0
2
0
0
.
0
5
5
0
.
2
9
3
0
.
0
3
1
?
0
.
1
4
5
0
.
0
5
4
0
.
5
3
3
?
0
.
1
6
1
?
0
.
3
1
5
?
0
.
0
3
0
0
.
0
2
2
?
0
.
1
7
8
I
N
T
C
O
V
?
0
.
0
5
7
?
0
.
1
7
6
0
.
1
1
1
?
0
.
0
7
3
0
.
1
3
6
?
0
.
0
1
0
0
.
0
0
5
0
.
2
4
8
0
.
0
4
8
?
0
.
1
9
7
?
0
.
0
5
1
?
0
.
0
5
6
?
0
.
0
0
6
0
.
0
3
0
B
O
N
U
S
0
.
1
4
1
?
0
.
0
3
1
0
.
0
7
3
0
.
3
1
0
0
.
1
3
9
?
0
.
1
2
0
0
.
0
3
7
0
.
2
4
7
0
.
0
1
0
?
0
.
0
3
9
?
0
.
2
0
3
?
0
.
0
4
2
0
.
0
0
3
?
0
.
0
9
7
C
E
O
O
W
N
?
0
.
1
9
1
0
.
0
3
4
0
.
0
6
7
?
0
.
0
6
1
?
0
.
0
0
8
?
0
.
1
4
8
?
0
.
1
2
8
?
0
.
1
2
5
0
.
0
6
4
?
0
.
0
5
6
0
.
1
1
8
?
0
.
0
9
7
?
0
.
0
4
7
?
0
.
0
1
4
I
N
S
T
p
0
.
1
2
4
0
.
0
9
6
0
.
0
1
0
0
.
1
3
7
?
0
.
1
8
5
0
.
0
7
0
?
0
.
2
1
6
?
0
.
0
1
5
?
0
.
1
4
0
0
.
3
5
1
?
0
.
0
9
3
?
0
.
0
3
5
0
.
0
2
0
?
0
.
1
2
7
S
I
Z
E
0
.
1
5
5
?
0
.
1
9
6
0
.
0
1
9
0
.
1
0
9
0
.
3
2
7
0
.
4
3
6
?
0
.
4
3
2
0
.
1
2
2
?
0
.
0
0
6
?
0
.
0
6
0
?
0
.
1
6
3
?
0
.
2
9
8
0
.
0
2
1
?
0
.
0
2
7
P
R
O
F
I
T
p
?
0
.
0
7
2
0
.
0
9
4
?
0
.
0
5
0
0
.
0
6
3
?
0
.
0
2
1
?
0
.
1
0
5
0
.
1
6
3
?
0
.
0
7
8
?
0
.
1
3
8
?
0
.
1
0
4
0
.
0
1
6
0
.
0
2
4
0
.
0
0
8
0
.
0
2
7
N
I
_
P
F
0
.
1
0
9
0
.
1
7
3
?
0
.
0
7
7
0
.
0
9
3
?
0
.
3
9
5
?
0
.
0
0
3
?
0
.
1
9
2
0
.
7
2
3
0
.
0
0
2
?
0
.
0
3
0
?
0
.
0
1
7
?
0
.
0
1
1
?
0
.
0
1
4
?
0
.
0
5
2
B
S
V
O
L
?
0
.
2
6
3
0
.
1
2
0
?
0
.
0
1
5
?
0
.
3
4
7
?
0
.
0
5
1
?
0
.
1
9
6
0
.
2
8
2
?
0
.
2
8
7
?
0
.
3
1
3
0
.
1
9
8
?
0
.
2
0
9
0
.
1
8
4
0
.
0
4
9
0
.
3
2
6
M
T
B
-
1
?
0
.
0
8
0
0
.
1
2
0
0
.
0
7
6
?
0
.
0
8
5
?
0
.
0
3
9
?
0
.
0
7
7
0
.
1
6
0
?
0
.
0
4
4
?
0
.
2
4
7
0
.
1
3
4
?
0
.
0
2
8
0
.
2
2
9
?
0
.
0
3
5
0
.
0
8
8
A
T
0
.
0
5
1
?
0
.
0
3
2
0
.
0
0
4
0
.
0
3
1
0
.
0
5
6
0
.
0
3
1
?
0
.
0
2
1
?
0
.
0
4
4
0
.
0
3
4
?
0
.
0
5
8
?
0
.
0
6
2
0
.
0
3
7
?
0
.
0
5
9
0
.
0
8
7
S
K
E
W
?
0
.
1
3
2
?
0
.
0
0
9
0
.
0
4
3
?
0
.
1
6
1
0
.
1
0
9
?
0
.
0
6
6
0
.
0
4
2
?
0
.
1
2
3
?
0
.
0
6
6
0
.
0
2
7
?
0
.
1
2
2
0
.
2
2
1
0
.
0
8
8
0
.
0
6
0
N
o
t
e
s
:
A
l
l
v
a
r
i
a
b
l
e
s
a
r
e
a
s
d
e
s
c
r
i
b
e
d
i
n
T
a
b
l
e
I
;
c
o
r
r
e
l
a
t
i
o
n
s
i
n
i
t
a
l
i
c
s
a
r
e
n
o
t
s
i
g
n
i
f
c
a
n
t
a
t
t
h
e
1
0
p
e
r
c
e
n
t
l
e
v
e
l
ARJ
27,2
138
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
that because ESO-expensing frms tended to have lower MTB-1 ratios, they had a
“buffer” in net assets for recording subsequent losses, which allows a frm greater
ability to write down assets. This indirectly increases AT by AT’s mechanical
connection to MTB-1. Another explanation is that frms with low MTB-1 ratios are
likely to get more pressure fromauditors to write down their assets (Kinney and Uecker,
1982; Pae et al., 2005), again indirectly increasing AT. Further, these results add to the
observed negative relationship between the MTB-1 ratio and AT in prior research
Table IV.
Likelihood of voluntary
SFAS 123 expense
recognition
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient (3) Coeffcient
Intercept ?5.095*** (37.74) ?3.672*** (16.77) ?5.116*** (23.49)
ISSUEp ? 2.237* (3.58) 1.966 (2.63) 2.782* (3.77)
AQUp ? ?9.300*** (9.92) ?9.378*** (9.92) ?8.100*** (6.63)
DEBTEQ ? ?0.010 (0.01) 0.036 (0.12) 0.112 (1.10)
INTCOV ? 0.134 (0.24) 0.050 (0.03) 0.169 (0.33)
BONUS ? 0.663* (3.39) 0.813** (4.84) 0.631 (2.53)
CEOOWN ? ?2.284 (1.55) ?1.521 (0.69) 0.430 (0.05)
INSTp ? 0.023 (0.00) 0.228 (0.21) 0.826 (1.96)
SIZE ? 0.298*** (22.21) 0.277*** (17.37) 0.321*** (19.23)
PROFITp ? 4.382* (3.37) 2.460 (0.91) 2.798 (1.03)
NI_PF ? ?33.534** (4.43) ?28.776* (3.22) ?6.444 (0.12)
NI_PFxINSTp ? 35.165* (3.66) 30.431* (2.75) 3.084 (0.02)
BSVOL ? ?2.570*** (15.23) ?2.544*** (12.87)
MTB-1 ? ?0.082* (3.77)
AT ? 0.068*** (7.66)
SSKEW ?
Number of ESO-expensing
frms 240 240 189
Number of ESO disclosure
frms 1,534 1,531 1,275
McFadden pseudo R-squared 18.6% 20.0% 15.7%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the
numbers in brackets belowthe coeffcients are Wald chi-squares; coeffcients are fromlogit regressions
of an indicator variable denoting whether, for fscal year 2004, the frm voluntarily expensed ESOs on
the fnancial statements under SFAS 123 over disclosure only; there were 155 ESOexpense-recognizing
frms for fscal year 2004 that were in the S&P 1500 index; the number of expense-recognizing frms is
less than the potential 2 years ? 155 ? 310 due to missing data in ExecuComp and Compustat; ten
industry dummy variables are not shown for ease of presentation; McFadden pseudo R-squared values
are calculated using a spreadsheet developed by Glynn (2003); all variables are winsorized at 1 and 99
per cent of their respective distributions and are described in Table I
139
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
(Pae et al., 2005; Beaver and Ryan, 2005; Roychowdhury and Watts, 2007; Lawrence
et al., 2012). Specifcally, a cleanly conservative accounting treatment of voluntarily
expensing ESOs on the fnancial statements is related differently to the MTB-1 ratio and
conditional conservatism, supporting Givoly et al.’s (2007) suggestion that relying on
any single proxy to assess a frm’s overall degree of conservatism could lead to
erroneous inferences. Although, there is a negative relationship between ATand MTB-1
ratio in this study, this may not be the case in other situations. The goal of this study is
to show how an actual accounting choice is related to AT and MTB-1.
Finally, the coeffcient [?
13
from equation (1)] on variable BSVOL in Table IV
(Column 2) is negative with a value ? ?2.570 and statistically signifcant at the 1 per
cent level, indicating that frms were less likely to voluntarily expense ESOs if their
returns had greater volatility. This supports the prediction that frms were concerned
with howmuch and howvariable their ESOexpense would be, and, arguably, this is the
most important variable in explaining the decision to voluntarily expense ESOs[17].
Similar values and signifcance levels are obtained for BSVOL using the ATmeasure of
conservatism in the main model (Table IV, Column 3).
7. Additional analysis and robustness tests of the main results
7.1 AT and controlling for MTB-1 ratio
Pae et al. (2005) recommend controlling for the MTB-1 ratio when using the ATmeasure
to proxy for conservatism, and, more recently, Lawrence et al. (2012) also support the use
of controlling for the MTB-1 ratio. Table V(Column 1) presents fndings fromestimating
equation (1), for fscal years 2003 and 2004, using both the MTB-1 ratio and ATmeasure
in the same logistic regression. The coeffcient on MTB-1 is ?0.057 and the coeffcient
on AT is 0.067, statistically insignifcant and signifcant at the 1 per cent level,
respectively. Therefore, controlling for the MTB-1 supports the main quantitative and
qualitative results regarding conditional conservatism.
7.2 Third proxy for conservatism
Prior literature suggests using multiple proxies for conservatism, as there is not a
generally accepted measure for conservatism (Givoly et al., 2007; Beatty et al., 2008;
Hui et al., 2009). In Table V (Column 2) are the regression results of equation (1) with
the conservatism proxy SKEW. SKEW is measured as the difference between cash
fows (COMPUSTAT #308) and earnings (COMPUSTAT #172) defated by total
assets (COMPUSTAT #6) summed over the prior fve fscal years, similar to Beatty
et al. (2008). This measure is selected because arguably it contains items related to
both the MTB-1 ratio and conditional conservatism. Accounting earnings, which are
part of SKEW, are determined using mandatory accounting standards (the MTB-1
ratio) and include decisions made by the frm to include losses in a timelier manner
than gains (conditional conservatism)[18].
The coeffcient on SKEW is ?0.224 but statistically insignifcant at the 10 per cent
level. This fnding is consistent with the results in Table IV of the MTB-1 ratio and
conditional conservatism working in opposite directions for the likelihood of ESO
expense recognition, as SKEW is related to both proxies of conservatism.
7.3 Exclusion of fnancial frms
Many fnancial services frms simultaneously announced their intention to recognize
ESO expenses on the fnancial statements. This is evidenced from the large proportion
ARJ
27,2
140
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
of frms from this industry in the expense recognition sample from both this and the
Aboody et al. (2004) study. Such an industry-wide decision to expense ESOs could refect
factors other than those included in the model. Financial institutions also differ fromother
frms in their debt to equity relationship, and Aboody et al. (2004) note that they likely have
lower stock-based compensation expense and may have a different investor base.
Table VI presents the pooledcross-sectional logistic regressionfndings fromestimating
equation (1), for fscal years 2003 and 2004, after eliminating fnancial service frms. The
coeffcient on MTB-1 is ?0.070 and still statistically signifcant at the 10 per cent level
(Column 1). The R
2
decreases from20.2 per cent to 15.5 per cent, likely due to the decrease in
Table V.
Likelihood of voluntary
SFAS 123 expense
recognition-alternative
proxies
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient
Intercept ?5.053*** (23.01) ?3.790*** (16.60)
ISSUEp ? 2.629*(3.31) 2.224* (3.24)
AQUp ? ?8.090*** (6.66) ?9.494*** (9.52)
DEBTEQ ? 0.134 (1.58) 0.032 (0.09)
INTCOV ? 0.075 (0.06) 0.016 (0.00)
BONUS ? 0.658* (2.75) 0.766** (4.06)
CEOOWN ? 0.487 (0.07) ?2.241 (1.39)
INSTp ? 0.837 (2.02) ?0.013 (0.00)
SIZE ? 0.341*** (20.66) 0.251*** (14.43)
PROFITp ? 2.256 (0.68) 4.623* (2.88)
NI_PF ? ?5.690 (0.10) ?49.565*** (7.76)
NI_PFxINSTp ? 3.324 (0.03) 59.598*** (7.92)
BSVOL ? ?2.596*** (13.17) ?2.409*** (12.84)
MTB-1 ? ?0.057 (1.57)
AT ? 0.067*** (7.52)
SSKEW ? ?0.224 (0.29)
Number of ESO-expensing frms 189 235
Number of ESO disclosure frms 1,273 1,508
McFadden pseudo R-squared 20.2% 20.8%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the numbers
inbrackets belowthe coeffcients are Waldchi-squares; coeffcients are fromlogit regressions of anindicator
variable denoting whether, for fscal year 2004, the frm voluntarily expensed ESOs on the fnancial
statements under SFAS 123 over disclosure only; there were 155 ESO expense-recognizing frms for fscal
year 2004that were inthe S&P1500index; the number of expense-recognizingfrms is less thanthe potential
2 years ?155 ?310 due to missing data in ExecuComp and Compustat; ten industry dummy variables are
not shown for ease of presentation; McFadden pseudo R-squared values are calculated using a spreadsheet
developed by Glynn (2003); all variables are winsorized at 1 and 99 per cent of their respective distributions
and are described in Table I
141
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
sample size. In addition, the results from Columns 2 and 3, using AT and SKEW as
conservatism proxies, also support the main fndings. In summary, the fnancial services
industry does not affect the main quantitative and qualitative results[19].
7.4 Firms that voluntarily announced expensing of ESOs
Determination of the frms that voluntarily announced plans to expense ESOs is
accomplished by collecting voluntary ESO (Statement of Financial Accounting
Standards, SFAS 123, 1995) expense recognition announcements from press
Table VI.
Likelihood of SFAS 123
expense recognition-
excluding fnancial
services frms
Panel A Sample period: 2003-2004
Variable Prediction (1) Coeffcient (2) Coeffcient (3) Coeffcient
Intercept ?1.935* (3.57) ?2.700** (5.49) ?2.329** (4.67)
ISSUEp ? 0.824 (0.25) 0.636 (0.10) 0.880 (0.25)
AQUp ? ?6.102** (4.75) ?4.437 (2.22) ?6.156** (4.20)
DEBTEQ ? ?0.058 (0.22) 0.041 (0.11) ?0.043 (0.11)
INTCOV ? 0.088 (0.08) 0.284 (0.91) ?0.067 (0.03)
BONUS ? 0.524 (1.22) 0.148 (0.09) 0.572 (1.27)
CEOOWN ? ?3.489 (2.49) ?1.836 (0.66) ?4.337* (3.12)
INSTp ? ?2.284*** (12.53) ?2.389*** (10.32) ?2.335*** (12.06)
SIZE ? 0.252*** (11.31) 0.289*** (12.78) 0.229*** (8.81)
PROFITp ? 8.287*** (7.49) 9.424*** (8.36) 10.706*** (10.56)
NI_PF ? ?24.991* (3.02) ?15.207 (0.65) ?44.118*** (5.65)
NI_PFxINSTp ? 7.814 (0.25) ?3.689 (0.03) 33.734 (2.18)
BSVOL ? ?0.965 (1.83) ?0.930 (1.48) ?0.920 (1.45)
MTB-1 ? ?0.070* (2.78)
AT ? 0.067*** (5.62)
SSKEW ? 0.055 (0.01)
Number of ESO-expensing frms 143 121 140
Number of ESO disclosure frms 1,427 1,199 1,406
McFadden pseudo R-squared 15.5% 17.1% 15.7%
Notes:
EXPD
it
? ?
0
? ?
1,N ?
1
N
IND
Ni
? ?
2
ISSUEp
it
? ?
3
AQUp
it
? ?
4
DEBTEQ
it
? ?
5
INTCOV
it
? ?
6
BONUS
it
? ?
7
CEOOWN
it
? ?
8
INSTp
it
? ?
9
SIZE
it
? ?
10
PROFITp
it
? ?
11
NI_PF
it
? ?
12
NI_PFxINSTp
it
? ?
13
BSVOL
it
? ?
14
?
MTB
? 1, AT, SKEW?
it ? ?
it
;
*; **; and ***denote signifcance at the 10, 5 and 1 per cent levels, two-tailed, respectively; the
numbers in brackets below the coeffcients are Wald chi-squares; coeffcients are from logit
regressions of an indicator variable denoting whether, for fscal year 2004, the frm voluntarily
expensed ESOs on the fnancial statements under SFAS 123 over disclosure only; there were 155
ESO expense-recognizing frms for fscal year 2004 that were in the S&P 1500 index; the number of
expense-recognizing frms is less than the potential 2 years. ? 155 ? 310 due to missing data in
ExecuComp and Compustat; ten industry dummy variables are not shown for ease of presentation;
McFadden pseudo R-squared values are calculated using a spreadsheet developed by Glynn (2003);
all variables are winsorized at 1 and 99 per cent of their respective distributions and are described
in Table I
ARJ
27,2
142
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
releases in the LexisNexis news retrieval service using search terms “stock options”
and “expense”. In addition, a signifcant portion of these companies were obtained
from the ESO announcing frms listed in Ciesielski (2004). The search in LexisNexis
was conducted for announcements from January 2002 through December 31, 2003,
from which 146 frms were obtained. Of these frms, 40 were either not included in
Compustat or not in the S&P 1500 index. Aboody et al. (2004) identify 155 frms that
announced plans to recognize the expense in 2002 and early 2003. During 2003 and
2004, few frms announced their intention to expense ESOs.
In untabulated results, using frms that announced they would be expensing ESOs,
provided results consistent with using frms that actually expensed ESOs during the
SFAS 123 period. The MTB-1 ratio is still negatively related and the AT measure still
positively related to frms’ likelihood of expensing ESOs during the voluntary period of
SFAS 123 at statistically signifcant levels.
7.5 Beginning-of-year AT and SKEW
In the main model, the MTB-1 variable is the beginning-of-period value, but the
other two conservatism proxies, AT and SKEW, are the end-of-period values. Many
frms’ decision to expense ESOs could have been made based on their
beginning-of-period (ex-ante) degree of conservatism. Equation (1) is run again, but
replaced with the beginning-of-period AT and SKEW separately. In untabulated
results, AT is still positively related to EXPD and SKEW is still statistically
insignifcant. Therefore, the main fndings are the same when using
beginning-of-period conservatism proxies.
8. Summary and conclusions
This study examines the relationship between frms’ decisions to expense ESOs under
the voluntary period of SFAS 123 and their beginning-of-period MTB-1 ratio and
conditional conservatism. Even though expensing of ESOs is a conservative accounting
treatment of an economic activity, it is not obvious fromprior literature howthe decision
would be related to the MTB-1 ratio and conditional conservatism.
During the period when expensing ESOs was voluntary, SFAS 123, the MTB-1 ratio
was less likely associated with ESO expense recognition, but conditional
conservatismwas more likely associated with ESOexpense recognition. The MTB-1
ratio fndings are consistent with the incentive and asymmetric information
explanations of conservatism, reducing the need for additional voluntary
information. Regarding conditional conservatism, results are consistent with
Roychowdhury and Watts’ (2007) term of a “buffer” in net assets for recording
subsequent losses, and Beaver and Ryan’s (2005) analytical relationship that shows
the MTB-1 ratio preempting conditional conservatism. These fndings add to the
literature on the negative relationship between the MTB-1 ratio and conditional
conservatism and show that conservatism plays a role in the disclosure versus
recognition decision. Furthermore, the results support using more than one measure
in studies that involve accounting conservatism because each proxy of
conservatism can represent different forms and motives of conservatism.
This study makes a connection between conservatism and a frm’s willingness to
incur real costs from their fnancial reporting. Many frms lobbied against
expensing stock options, refecting the widely held perception that there were
143
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
substantial costs in recognizing ESOs within the fnancial statements. In addition,
this study shows that frms took into account the amount and variability of the ESO
expense because stock return variability is negatively related to ESO expense
recognition during the SFAS 123 period.
As with all studies related to accounting conservatism, there is the limitation of
suitable proxies for conservatism. Caution is advised to not interpret the results of this
study as a general relationship between the MTB-1 ratio and conditional conservatism
because this study is over a period of only two years; the frms are taken from the S&P
1500 index; and the relationship of the MTB-1 ratio and conditional conservatism is
specifc to ESO expense recognition.
Notes
1. Firms’ overall or general conservatismcan be described as the degree to which accounting
decisions lower income. By this description, the decision to expense ESOs is a
discretionary conservative choice because it clearly decreases income. Aboody et al.
(2004, p. 128) state “Recognition of SFAS 123 expense is a conservative accounting choice
consistent with that modeled in Hughes and Levine (2003) because it unambiguously
lowers net income relative to the disclosure-only alternative”. However, it is not evident
howMTB-1 and conditional conservatismwould be related to this decision because of the
multifaceted nature of the MTB-1 ratio. For example, it can also represent growth frms
(Fama and French, 1992).
2. Voluntarily expensing of ESOs is determined by “OE” in Compustat Mnemonic STKOF,
which represents frms that were early adopters of SFAS 123R.
3. Even non-academics like the Association of Investment Management and Research have
stated repeatedly over the years that stock options should be expensed and strongly
supported the 1994 proposal of the Financial Accounting Standards Board (FASB) to
require companies to expense stock options in the income statement. Nevertheless,
prominent business leaders, e.g. Andrew Grove of Intel and Harvey Golub, formerly of
American Express, argued against expensing of ESOs on the fnancial statements (Guay
et al., 2003).
4. Conservatism is defned as early as 1924 by Bliss (1924, p. 110). His defnition included
“anticipate no profts, but anticipate all losses”. However, FAS?B’s Conceptual Framework
for Financial Reporting (2010) does not include the principle of accounting conservatism
because it distorts the neutralityof fnancial information. Nevertheless, conservatismremains
prevalent in practice (Lawrence et al., 2012).
5. Even though the MTB ratio is related to unconditional conservatism, this study does not
imply this proxy of conservatism to mean unconditional conservatism.
6. Option values are highly dependent and positively related to a frm’s stock return
volatility (Merton, 1973; Hull, 2006, Chapters 15, 16, and 19). Blacconiere et al. (2011)
examine the voluntary disclosure of the reliability of stock option compensation during
the SFAS 123 period. They fnd stock return volatility to be positively related to this type
of voluntary disclosure.
7. It is possible that some frms will expense ESOs because they expect it to become mandatory
(Schrand, 2004). However, when frms that expensed during fscal year 2003 are used, this did
not change the main fndings of the study. Both proxies, MTB-1 and AT, have the same sign
ARJ
27,2
144
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
andare signifcant at the 1 and10 per cent level, respectively. SKEWis still insignifcant at the
10 per cent level. The sample periodis not extendedbeyondtwo years because a frm’s general
degree of conservatism is likely not constant for long periods.
8. For additional testing, frms that announced that they would be expensing voluntarily
ESOs are also examined (Section 7.4). Determination of the frms that voluntarily intend
to expense ESOs is accomplished by collecting voluntary ESO (SFAS 123) expense
recognition announcements from press releases on LexisNexis news retrieval service
using search terms “stock options” and “expense”. In addition, a signifcant portion of
voluntarily expensing companies was obtained from the ESO-expensing companies
listed in Ciesielski (2004).
9. The model of Aboody et al. (2004) is reproduced below, where the dependent variable
RECOGNIZEis 1 if a frmvoluntarily expensed ESOs and 0 otherwise. OSDIRSTK(outstanding
director stock) is not included because the data are not in the ExecuComp data set that was
available and also exclude %OPTTOP5 (per cent options held by top fve paid executives). The
latter is excluded because including it in any of the logistic regressions causes the validity of the
model tobe questionable andthe MLEmaynot exist. Still, the questionable results whenincluding
%OPTTOP5 did not change the main fndings of this paper. In the Aboody et al. model,
%OPTTOP5 is statistically insignifcant with a p-value ?0.472.
RECOGNIZEj ? ?0 ?1
N
INDNj ? ?1ISSUEj ? ?2ACQj ? ?3DEBT Eqj ??4INT COVj ?
?5BONUSj ? ?6CEO OWNj ??7OSDIRSTKj ? ?8INSTITj ? ?9SIZEj ??10PROFITj ?
?11%OPT TOP 5 j ??12COMPEXj ??13COMPEXj ?INSTITj ??j.
The primary fndings of Aboody et al. (2004) are that all their variables, except the last four, are
statistically signifcant at the 10 per cent level or better.
10. The ten industry groupings used as dummy variables are Mining/Construction/Extractive,
Food, Textiles/Print/Publish, Chemicals, Manufacturing, Computers, Transportation,
Utilities/Financial/Insurance and Retail. The Services industry is used as the control group.
The tables do not include the industry dummies for ease of presentation. The full table is
available on request from the author.
11. Guay et al. (2003, p. 408) note that empirical evidence shows frms that most actively lobbied
against ESO expensing are characterized by top executives having a greater portion of their
compensation from options. It is possible that CEOOWN will be negatively related to ESO
expense recognition.
12. For missing values in ExecuComp, the standard deviation of the prior fve years’ returns is
used. ExecuComp calculates BSVOL as the standard deviation calculated over 60 months.
13. As ?
2
in the denominator can be a very small number leading to extreme values for AT, AT
is restricted to values from zero to ten.
14. For ESO-expensing frms: CEOinformation is hand-collected fromproxy statement DEF 14A
for missing data in ExecuComp and INTCOV information is obtained from S&P’s Capital IQ
for missing data in Compustat. For the two years tested, there are 3000 S&P 1500 frms. Of
these, only 2,687 have an MTB ratio within Compustat. Given that approximately 10 per cent
of the frms voluntarily expensed ESOs in 2004, this would mean that we have 269 expensing
frms that have a MTBratio. However, there were still some frms for which CEOinformation
from the DEF 14A proxy statement was not available. These are the main reasons the fnal
sample for expensing frms when using the MTB ratio is only 240 frms.
145
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
15. Acomparison to the Aboody et al. (2004) would not be very meaningful because they use fscal
year 2001 for examination and some variable defnitions are different to the ones used this
study. For example, ISSUEp and AQUp are indicator variables in the Aboody et al. model but
are continuous variables in this study. Also, the Aboody et al. paper uses frms that
announced they would expense ESOs. The main fndings of this paper use frms that actually
expensed ESOs during the voluntary SFAS 123 period.
16. The logistic regression with AT produced a Quasi-complete separation when using only two
years of observations for the complete logistic regression. I therefore performed Firth’s
penalized likelihood approach, which is a method of addressing issues of separability, small
sample sizes and bias of the parameter estimates. Heinze and Schemper (2002) fnd this
approach an “ideal solution to separation” for logistic regression. In any case, the coeffcients
and statistical signifcance are very similar for AT whether the Firth approach is used or not
for the regression including AT.
17. The McFadden pseudo R
2
is the second highest, at 15.7 per cent, when only using the industry
dummy variables and BSVOL. The only variable by itself which produces a higher McFadden
pseudo R
2
, at 16.5 per cent, is SIZE.
18. Many studies, for example Beatty et al. (2008), use a composite measure for conservatism. A
composite measure is also testedinthis studyusingequal weights of AT, MTB-1 andSKEW. The
method is to frst convert each of these conservatism proxies to deciles so they are given equal
weight and assign one-third weight to each of the proxy deciles. A minimum of two proxies is
required and if only two are available for a specifc frm-year, then [one-half] weight is assigned.
The composite measure, when used in the logistic regression of equation (1) as the conservatism
proxy, also gave statistically insignifcant results (p-value ?0.67). This was expected because the
AT and MTB-1 conservatism proxies were oppositely related to the decision to expense ESOs.
This study calculates ATas (?
2,it
??
3,it
/?
2,it
). ATis also tested using only ?
3
fromequation (2),
which is also used in Givoly et al. (2007). The coeffcient on the decile ranking of ?
3
when used to
determine its relationship to the decision to voluntarily expense stock options is 0.047 with a two
sided p-value of 0.115.
19. I also test the model using a matched size sample of expensing and non-expensing frms.
Again the results support the main fndings.
References
Aboody, D., Barth, M. and Kasznik, R. (2004), “Firms voluntary recognition of stock-based
compensation expense”, Journal of Accounting Research, Vol. 42 No. 2, pp. 123-150.
Accounting Principles Board Opinion No. 25 (1972), Accounting for Stock Issued to Employees,
APB, New York, NY.
Ahmed, A. and Duellman, S. (2007), “Accounting conservatism and board of director
characteristics: an empirical analysis”, Journal of Accounting and Economics, Vol. 43
Nos 2/3, pp. 411-437.
Ball, R. and Shivakumar, L. (2005), “Earnings quality in UK private frms: comparative loss
recognition timeliness”, Journal of Accounting and Economics, Vol. 39 No. 1, pp. 83-128.
Barth, M.E., Gow, I.D. and Taylor, D.J. (2012), “Why do pro forma and street earnings not refect
changes in GAAP? Evidence fromSFAS 123R”, Reviewof Accounting Studies, Vol. 17 No. 3,
pp. 526-562.
Basu, S. (1997), “The conservatism principle and the asymmetric timeliness of earnings”, Journal
of Accounting and Economics, Vol. 24 No. 1, pp. 3-37.
ARJ
27,2
146
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Beatty, A. (2007), “Discussion of asymmetric timeliness of earnings, market-to-book and
conservatism in fnancial reporting”, Journal of Accounting and Economics, Vol. 44 No. 1,
pp. 32-35.
Beatty, A., Weber, J. and Yu, J.J. (2008), “Conservatism and debt”, Journal of Accounting and
Economics, Vol. 45 Nos 2/3, pp. 154-174.
Beaver, W. and Ryan, S. (2005), “Conditional and unconditional conservatism, concepts and
modeling”, Review of Accounting Studies, Vol. 10 Nos 2/3, pp. 269-309.
Blacconiere, G., Frederickson, J., Johnson, M. and Lewis, M. (2011), “Are voluntary disclosures
that disavow the reliability of mandated fair value information informative or
opportunistic?”, Journal of Accounting and Economics, Vol. 52 Nos 2/3, pp. 235-251.
Black, F. and Scholes, M. (1973), “The pricing of options and corporate liabilities”, Journal of
Political Economy, Vol. 81 No. 3, pp. 637-654.
Bliss, J.H. (1924), Management Though Accounts, The Ronald Press Company, New York, NY.
Bodie, Z., Kaplan, R. and Merton, R.C. (2003), “For the last time, stock options are an expense”,
Harvard Business Review, Vol. 81 No. 3, pp. 63-73.
Botosan, C. and Plumlee, M. (2001), “Stock option expense: the sword of Damocles revealed”,
Accounting Horizons, Vol. 15 No. 4, pp. 311-328.
Ciesielski, J. (2004), “S&P 500 companies electing to expense stock option compensation”, The
Analyst’s Accounting Observer Weblog, January, 12, 2004, Subscriber only Blog.
Dietrich, J.R., Muller, K. A. III and Riedl, E.J. (2007), “Asymmetric timeliness tests of accounting
conservatism”, Review of Accounting Studies, Vol. 12 No. 1, pp. 95-124.
Fama, E. and French, K. (1992), “The cross-section of expected stock returns”, Journal of Finance,
Vol. 47 No. 2, pp. 427-465.
Feltham, J. and Ohlson, J. (1995), “Valuation and clean surplus accounting for operating and
fnancial activities”, Contemporary Accounting Research, Vol. 11 No. 2, pp. 689-731.
Fields, T., Lys, L. and Vincent, L. (2001), “Empirical research on accounting choice”, Journal of
Accounting and Economics, Vol. 31 No. 1/3, pp. 255-307.
Francis, J., LaFond, R., Olsson, P. and Schipper, K. (2004), “Costs of equity and earnings
attributes”, The Accounting Review, Vol. 79 No. 4, pp. 967-1010.
Gigler, B. and Hemmer, T. (2001), “Conservatism, optimal disclosure policy, and the timeliness of
fnancial reports”, The Accounting Review, Vol. 76 No. 4, pp. 471-493.
Givoly, D., Hayn, C. and Natarajan, A. (2007), “Measuring reporting conservatism”, The
Accounting Review, Vol. 82 No. 1, pp. 65-106.
Glynn, P. (2003), “Spreadsheet to calculate McFadden pseudo R
2
”, available at:
http://staff.washington.edu/glynn/r2pseudo.xls (accessed 11 November 2012).
Guay, W., Kothari, S.P. and Sloan, R. (2003), “Accounting for employee stock options”, The
American Economic Review, Vol. 92 No. 2, pp. 405-409.
Heinze, G. and Schemper, M. (2002), “A solution to the problem of separation in logistic
regression”, Statistics in Medicine, Vol. 21 No. 16, pp. 2409-2419.
Hsu, A., O’Hanlon, J. and Peasnell, K. (2012), “The Basu measure as an indicator of conditional
conservatism: evidence from UK earnings components”, European Accounting Review,
Vol. 21 No. 1, pp. 87-113.
Hughes, J.S. and Levine, C.B. (2005), “Management compensation and earnings-based covenants
in resolving adverse selection in credit markets”, Journal of Applied Corporate Finance,
Vol. 11, pp. 832-850.
147
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Hui, K.W., Matsunaga, S. and Morse, D. (2009), “The impact of conservatism on management
earnings forecasts”, Journal of Accounting and Economics, Vol. 47 No. 3, pp. 192-207.
Hull, J.C. (2006), Options, Futures and Other Derivatives, 6th ed., Prentice Hall, Upper Saddle River,
NJ.
Jenkins, D.S., Kane, G.D. and Velury, U. (2009), “Earnings conservatism and value relevance
across the business cycle”, Journal of Business Finance and Accounting, Vol. 39 Nos 9/10,
pp. 1041-1058.
Kinney, W. and Uecker, W. (1982), “Mitigating the consequences of anchoring in auditor
judgements”, The Accounting Review, Vol. 57 No. 1, pp. 55-69.
LaFond, R. and Watts, R. (2008), “The information role of conservatism”, The Accounting Review,
Vol. 83 No. 2, pp. 447-478.
Lawrence, A., Sloan, R. and Sun, Y. (2012), “Mandatorily conservative accounting, evidence and
implications”, working paper, U. C. Berkeley, Berkeley, CA.
Mason, L. (2004), “The impact of accounting conservatism on the magnitude of the differential
information content of cash fows and accruals”, Journal of Accounting, Auditing and
Finance, Vol. 19 No. 3, pp. 249-282.
Merton, R. (1973), “Theory of rational option pricing”, Bell Journal of Economics, Vol. 4 No. 1,
pp. 141-183.
Nasev, J. (2009), Conditional and Unconditional Conservatism Implications for Accounting Based
Valuation and Risky Projects, Gabler, Wiesbaden.
Pae, J., Thornton, D. and Welker, M. (2005), “The link between earnings conservatism and the
market-to-book ratio”, Contemporary Accounting Research, Vol. 22 No. 3, pp. 693-717.
Penman, S. and Zhang, X. (2002), “Accounting conservatism, the quality of earnings, and stock
returns”, The Accounting Review, Vol. 77 No. 2, pp. 237-264.
Qiang, X. (2007), “The effects of contracting, litigation, regulation, and tax costs on conditional
and unconditional conservatism: cross-sectional evidence at the frm level”, The
Accounting Review, Vol. 82 No. 3, pp. 759-796.
Roychowdhury, S. and Watts, R. (2007), “Asymmetric timeliness of earnings, market-to-book and
conservatismin fnancial reporting”, Journal of Accounting and Economics, Vol. 44 Nos 1/2,
pp. 2-31.
Rubinstein, M. (1995), “On the accounting valuation of employee stock options”, Journal of
Derivatives, Vol. 3 No. 1, pp. 8-24.
Ruddock, C., Taylor, S.J. and Taylor, S.L. (2006), “Nonaudit services and earnings conservatism: is
auditor independence impaired?”, Contemporary Accounting Research, Vol. 23 No. 3,
pp. 701-746.
Schrand, C. (2004), “Discussion of frms’ voluntary recognition of stock-based compensation
expense”, Journal of Accounting Research, Vol. 42 No. 2, pp. 151-158.
Statement of Financial Accounting Standards (SFAS) (1995), “Accounting for stock-based
compensation”, Summary of Statement No. 123, available at: www.fasb.org/st/
Watts, R.L. (2003a), “Conservatism in accounting part1: explanations and implications”,
Accounting Horizons, Vol. 17 No. 3, pp. 207-221.
Watts, R.L. (2003b), “Conservatism in accounting part II: evidence and research opportunities”,
Accounting Horizons, Vol. 17 No. 4, pp. 287-301.
Zhang, J. (2008), “The contracting benefts of accounting conservatismto lenders and borrowers”,
Journal of Accounting and Economics, Vol. 45 No. 1, pp. 27-54.
ARJ
27,2
148
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Further reading
Statement of Financial Accounting Standards Board (2004), “Share based payment”, Summary of
Statement No.123(R), available at: www.fasb.org/st/
About the author
Gulraze Wakil, BCOMM (U. of Toronto), MA (Economics-U. of Toronto), PhD, CGA received his
PhD in accounting (minor in fnance) from Kent State University, Ohio. He is an Assistant
Professor in the Sprott School of Business at Carleton University, Ottawa, Canada. His research
projects include accounting conservatism, accounting and capital markets, time-series forecasting
and international accounting. Teaching interests include all areas of accounting. He earned his
Certifed General Accountant Designation while working as an accountant in the insurance
industry in Toronto, Canada. Gulraze Wakil can be contacted at: [email protected]
To purchase reprints of this article please e-mail: [email protected]
Or visit our web site for further details: www.emeraldinsight.com/reprints
149
Market-to-book
ratio
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
:
2
0
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
doc_823226927.pdf