Description
association between institutional ownership
and income smoothing. The results support
the predicted positive association between
institutional ownership and the likelihood of
firms smoothing earnings towards their
earnings trend in general. However, this
association is not systematic across all firms.
The positive association is most evident among
profit firms with pre-managed earnings above
their earnings trend. No significant association
is found for profit firms with pre-managed
earnings below their earnings trend and loss
firms in general.
Accounting Research Journal
Institutional Ownership and Income Smoothing: Australian Evidence
Ping-Sheng Koh
Article information:
To cite this document:
Ping-Sheng Koh, (2005),"Institutional Ownership and Income Smoothing: Australian Evidence ", Accounting Research
J ournal, Vol. 18 Iss 2 pp. 93 - 110
Permanent link to this document:http://dx.doi.org/10.1108/10309610580000678
Downloaded on: 24 January 2016, At: 20:59 (PT)
References: this document contains references to 0 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 604 times since 2006*
Users who downloaded this article also downloaded:
Chi-Yih Yang, Boon Leing Tan, Xiaoming Ding, (2012),"Corporate governance and income smoothing in China", J ournal of
Financial Reporting and Accounting, Vol. 10 Iss 2 pp. 120-139http://dx.doi.org/10.1108/19852511211273688
Rusmin Rusmin, Glennda Scully, Greg Tower, (2012),"Income smoothing behaviour by Asian transportation firms",
Managerial Auditing J ournal, Vol. 28 Iss 1 pp. 23-44http://dx.doi.org/10.1108/02686901311282489
Ahsan Habib, Haiyan J iang, (2012),"Managerial ownership-induced income smoothing and information asymmetry", Pacific
Accounting Review, Vol. 24 Iss 2 pp. 211-232http://dx.doi.org/10.1108/01140581211259839
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Institutional Ownership and Income Smoothing: Australian Evidence
93
Institutional Ownership and Income
Smoothing: Australian Evidence
Ping-Sheng Koh
UQ Business School
University of Queensland
Abstract
This study examines the rarely investigated
association between institutional ownership
and income smoothing. The results support
the predicted positive association between
institutional ownership and the likelihood of
firms smoothing earnings towards their
earnings trend in general. However, this
association is not systematic across all firms.
The positive association is most evident among
profit firms with pre-managed earnings above
their earnings trend. No significant association
is found for profit firms with pre-managed
earnings below their earnings trend and loss
firms in general. This study also finds that, in
Australia, while institutional ownership has a
non-linear association with income increasing
earnings management (Koh, 2003), such
association manifests itself within the income
smoothing framework. The results of this study
highlight the complexities in the association
between institutional ownership and earnings
management strategies, and future research can
benefit by explicitly examining the trade-offs
between alternative earnings management
incentives and the factors that affect the relative
strength of these incentive trade-offs.
Keywords: Institutional Investors; Discretionary Accruals;
Earnings Management; Income Smoothing.
JEL classification: G32, M41
Acknowledgements: The author was a doctoral candidate at
the University of Tasmania when the background research to
this paper and initial data collection commenced. Comments
by Peter Clarkson, Phil Gray, Joseph Fan, Jayne Godfrey,
Terry O’Keefe, Norman Wong, and seminar participants at
the University of Tasmania, and participants at the 2002
AAANZ Conference and the 2002 EAA Annual Congress
are greatly appreciated. I acknowledge the financial support
from the UQ Business School.
1. Introduction
Despite growing numbers of studies on the
association between institutional ownership and
earnings management (e.g. Bushee, 1998;
Cheng and Reitenga, 2000; Koh, 2003), the
association between institutional ownership and
income smoothing is rarely investigated
(Carlson and Bathala, 1997).
1
This study
examines the association between institutional
ownership and the likelihood of a portfolio firm
smoothing reported earnings through accruals
management. Income smoothing in this study
refers to earnings management toward an
earnings trend to reduce earnings variability
(e.g., Beidleman, 1973).
2
The practice of
income smoothing is well established (e.g.,
Beidleman, 1973; Godfrey and Jones, 1999;
Ronen and Sadan, 1981; Subramanyam, 1996)
and firms are facing increasing pressure to
report smooth earnings. For example, a
prominent FORTUNE 500 firm’s CEO claims
that, “[t]he No. 1 job of management is to
smooth out earnings” (Loomis, 1999).
Managers have incentives to smooth reported
earnings as it can affect earnings predictability,
perceived risk of the firm, managers’ personal
wealth and job security, among others. That is,
1 Bushee (1998) examines earnings management via
cutting research and development expenditure; Cheng
and Reitenga (2000) examine accruals management;
whilst Koh (2003) examines income increasing accruals
management. Carlson and Bathala (1997) is the only
known study directly investigate the association
between institutional ownership and income smoothing.
There are no known studies on the association between
institutional ownership and income smoothing in
Australia.
2 For this study, a firm is smoothing income if
Abs(Earnings-Trend)0.10).
Spearman correlations between the
explanatory variables are documented in
Table 3. Total institutional ownership (PIO) is
positively associated with firm size (SIZE)
consistent with the popularity of indexing
strategies among institutional investors in
Australia (Stapledon, 1996a; 1996b). The
positive association between institutional
ownership and firm size is also consistent
with the U.S. evidence (e.g., Carlson and
Bathala, 1997). Some weak evidence that
institutional ownership is negatively associated
with leverage also exists (p0.10; Column 3 of
13 Incidentally, all profit (loss) firms have positive
(negative) non-discretionary earnings.
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ACCOUNTING RESEARCH JOURNAL VOLUME 18 NO 2 (2005)
104
Table 4
Logit Regression of Income Smoother on Institutional Ownership Variables
and Control Variables
Prob(SMOOTH
= 1) = ? + ?
1
PIO + ?
2
SIZE + ?
3
LEV + ?
4
MGRSH + ?
5
CFO + ?
6
MINE +
!
=
1997
1994 k
?
k
YEAR
k
Estimated Coefficients (p-value)*
Variable Expected Sign
Full Sample Profit Firms
#
Loss Firms
##
(1) (2) (3)
Constant ? 2.1074
(0.0641)
2.1109
(0.1101)
2.1490
(0.5556)
PIO + 1.4505
(0.0283)
2.2064
(0.0068)
-2.1141
(0.1687)
SIZE +/- -0.1579
(0.1514)
-0.2445
(0.0638)
0.2513
(0.4838)
LEV - -0.8587
(0.0579)
-0.3280
(0.3409)
-1.2319
(0.1390)
MGRSH + 0.6729
(0.2290)
1.0316
(0.1620)
-3.3449
(0.1068)
CFO - -2.0895
(0.0795)
-1.0267
(0.3018)
-6.0148
(0.0620)
MINE +/- -0.2946
(0.4824)
-0.0504
(0.9227)
-0.7425
(0.6515)
Nagelkerke R
2
0.0980 0.1329 0.4734
Sample size (N) 202 156 46
* 1-tail p-value is reported if direction of the coefficient is predicted, otherwise 2-tail p-value is reported
# Profit firms all have NDE > 0
## Loss firms all have NDE < 0
PIO = Percentage institutional ownership
SIZE = Natural logarithm of total assets
LEV = Ratio of total liabilities to total tangible assets
MGRSH = Ratio of directors’ shareholdings (direct, indirect and beneficial holdings) to total ordinary shares outstanding
CFO = Cash flows from operations/total assets
MINE = Mining company dummy variables (1 if a firm is a mining firm; 0 otherwise)
SMOOTH = 1 if a firm an income smoother, i.e., Abs(Earnings – Trend) < Abs(NDE – Trend); 0 otherwise
Earnings = Current year earnings before interests and tax and before extraordinary items divided by prior year total assets
Trend = Prior year earnings before interests and tax and before extraordinary items divided by prior year total assets
NDE = Non-discretionary earnings as measured by earnings less discretionary accruals (DACC) as proxied by
modified Jones accruals model
Table 4). These findings are consistent with
existing argument that earnings management
incentives, including income smoothing
incentives, are likely to be stronger for profit
than loss firms, and hence their earnings
management strategies are likely to differ.
4.3 Are Income Smoothing Incentives
Uniform Across All Circumstances?
As posited in Section 2.2, it is also expected
that the predicted positive association between
institutional ownership and income smoothing
is more prevalent among firms with pre-
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Institutional Ownership and Income Smoothing: Australian Evidence
105
Table 5
Logit Regression of Income Smoother on Institutional Ownership Variables
and Control Variables (Additional Tests)
Prob(SMOOTH
= 1) = ? + ?
1
PIO + ?
2
SIZE + ?
3
LEV + ?
4
MGRSH + ?
5
CFO + ?
6
MINE +
!
=
1997
1994 k
?
k
YEAR
k
Estimated Coefficients (p-value)*
Constant PIO SIZE LEV MGRSH CFO MINE Nagelkerke
? + +/- - + - +/- R
2
Panel A: Firms with NDE above their earnings trend (n=110)
1.2486
(0.4211)
3.0255
(0.0034)
-0.1732
(0.2902)
-0.0769
(0.4647)
2.0358
(0.0620)
0.7385
(0.3750)
-0.6730
(0.2460)
0.1810
Panel B: Firms with NDE below their earnings trend (n=92)
2.0665
(0.3091)
0.3867
(0.3850)
-0.0185
(0.9203)
-1.9516
(0.0426)
-1.2823
(0.2153)
-5.0078
(0.0251)
0.3016
(0.6977)
0.2384
Panel C: Profit firms with NDE above their earnings trend (n=102)
0.8587
(0.5935)
3.1538
(0.0036)
-0.1848
(0.2844)
0.2459
(0.4019)
2.4865
(0.0457)
1.2061
(0.3130)
-0.4577
(0.4973)
0.2170
Panel D: Profit firms with NDE below their earnings trend (n=54)
9.3702
(0.7124)
1.2587
(0.2585)
-0.3200
(0.2438)
-1.7281
(0.1763)
-1.6635
(0.1901)
-8.4417
(0.0386)
0.9215
(0.4035)
0.4053
Panel E: Firms with positive discretionary accruals (n=105)
0.7780
(0.6274)
2.0976
(0.0397)
-0.2741
(0.1223)
0.2021
(0.4161)
-0.5918
(0.3439)
-5.3352
(0.0092)
0.7224
(0.2477)
0.1900
*1-tail p-value is reported if direction of the coefficient is predicted, otherwise 2-tail p-value is reported
PIO = Percentage institutional ownership
SIZE = Natural logarithm of total assets
LEV = Ratio of total liabilities to total tangible assets
MGRSH = Ratio of directors’ shareholdings (direct, indirect and beneficial holdings) to total ordinary shares outstanding
CFO = Cash flows from operations/total assets
MINE = Mining company dummy variables (1 if a firm is a mining firm; 0 otherwise)
SMOOTH = 1 if a firm an income smoother, i.e., Abs(Earnings – Trend) < Abs(NDE – Trend); 0 otherwise
Earnings = Current year earnings before interests and tax and before extraordinary items divided by prior year total assets
Trend = Prior year earnings before interests and tax and before extraordinary items divided by prior year total assets
NDE = Non-discretionary earnings as measured by earnings less discretionary accruals (DACC) as proxied by
modified Jones accruals model
managed earnings above their earnings trend
(i.e., NDE > Trend).
14
To investigate this, the
logit regression is regressed on two sub-
samples, viz., firms with pre-managed earnings
(NDE) above versus below their earnings trend
14 To the extent that firms with NDE greater than their
earnings trend do not manage earnings downwards to
smooth earnings, it would bias against finding results
consistent with this expectation.
(results are reported in Panels A and B of Table
5 respectively).
15
For firms with NDE above
their earnings trend, the estimated coefficient of
the institutional ownership is positive and
significant (p=0.0034), while firms with NDE
15 Income smoothers among firms with pre-managed
earnings above (below) their earnings trend correspond
to firms that smooth their earnings down (up) towards
their earnings trend.
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ACCOUNTING RESEARCH JOURNAL VOLUME 18 NO 2 (2005)
106
below their earnings trend have insignificant
coefficient (p=0.3850). These results suggest
that firms with high institutional ownership are
more likely to smooth income if they
are above their earnings trend before
accruals management. In contrast, institutional
ownership is not associated with income
smoothing when firms’ NDE is below their
earnings trend. Therefore, the initial findings
on Column 1 of Table 4 can be attributed
mainly to the association between institutional
ownership and income smoothing among firms
with pre-managed earnings above their
earnings trend.
The above findings also provide preliminary
empirical evidence indicating that the choices
between income smoothing and other earnings
management strategies that a manager faces
will depend upon whether a firm’s pre-
managed earnings is higher or lower than its
earnings trend. The results suggest that such
incentive trade-offs are less ambiguous among
firms with pre-managed earnings that exceed
their earnings trend than those below the trend.
Income Smoothing Incentives of Profit
versus Loss Firms
Further tests are conducted to examine whether
the differences between firms with NDE above
and below their earnings trend persist when
only profit firms are considered. Consistent
with the preceding findings, for profit firms
with NDE that exceed their earnings trend,
institutional ownership (PIO) is significantly
positively associated with income smoothing at
the 1% level (Table 5, Panel C). However, for
profit firms with NDE below their earnings
trend, their institutional ownership is not
associated with income smoothing (Table 5,
Panel D).
Combining the results from Panels B and D
of Table 5 indicates that for firms below their
earnings trend prior to accruals management,
institutional ownership is not associated with
income smoothing regardless of whether they
are profit or loss making firms.
16
16 For brevity, the results for loss firms with NDE below
or above their earnings trend are not reported. These
results reveal that institutional ownership is not
associated with income smoothing (p-values>0.10),
consistent with earlier findings.
Among the control variables, cash flows
from operations (CFO) are negatively
associated with income smoothing only among
firms with NDE below their earnings trend (p-
values
association between institutional ownership
and income smoothing. The results support
the predicted positive association between
institutional ownership and the likelihood of
firms smoothing earnings towards their
earnings trend in general. However, this
association is not systematic across all firms.
The positive association is most evident among
profit firms with pre-managed earnings above
their earnings trend. No significant association
is found for profit firms with pre-managed
earnings below their earnings trend and loss
firms in general.
Accounting Research Journal
Institutional Ownership and Income Smoothing: Australian Evidence
Ping-Sheng Koh
Article information:
To cite this document:
Ping-Sheng Koh, (2005),"Institutional Ownership and Income Smoothing: Australian Evidence ", Accounting Research
J ournal, Vol. 18 Iss 2 pp. 93 - 110
Permanent link to this document:http://dx.doi.org/10.1108/10309610580000678
Downloaded on: 24 January 2016, At: 20:59 (PT)
References: this document contains references to 0 other documents.
To copy this document: [email protected]
The fulltext of this document has been downloaded 604 times since 2006*
Users who downloaded this article also downloaded:
Chi-Yih Yang, Boon Leing Tan, Xiaoming Ding, (2012),"Corporate governance and income smoothing in China", J ournal of
Financial Reporting and Accounting, Vol. 10 Iss 2 pp. 120-139http://dx.doi.org/10.1108/19852511211273688
Rusmin Rusmin, Glennda Scully, Greg Tower, (2012),"Income smoothing behaviour by Asian transportation firms",
Managerial Auditing J ournal, Vol. 28 Iss 1 pp. 23-44http://dx.doi.org/10.1108/02686901311282489
Ahsan Habib, Haiyan J iang, (2012),"Managerial ownership-induced income smoothing and information asymmetry", Pacific
Accounting Review, Vol. 24 Iss 2 pp. 211-232http://dx.doi.org/10.1108/01140581211259839
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Institutional Ownership and Income Smoothing: Australian Evidence
93
Institutional Ownership and Income
Smoothing: Australian Evidence
Ping-Sheng Koh
UQ Business School
University of Queensland
Abstract
This study examines the rarely investigated
association between institutional ownership
and income smoothing. The results support
the predicted positive association between
institutional ownership and the likelihood of
firms smoothing earnings towards their
earnings trend in general. However, this
association is not systematic across all firms.
The positive association is most evident among
profit firms with pre-managed earnings above
their earnings trend. No significant association
is found for profit firms with pre-managed
earnings below their earnings trend and loss
firms in general. This study also finds that, in
Australia, while institutional ownership has a
non-linear association with income increasing
earnings management (Koh, 2003), such
association manifests itself within the income
smoothing framework. The results of this study
highlight the complexities in the association
between institutional ownership and earnings
management strategies, and future research can
benefit by explicitly examining the trade-offs
between alternative earnings management
incentives and the factors that affect the relative
strength of these incentive trade-offs.
Keywords: Institutional Investors; Discretionary Accruals;
Earnings Management; Income Smoothing.
JEL classification: G32, M41
Acknowledgements: The author was a doctoral candidate at
the University of Tasmania when the background research to
this paper and initial data collection commenced. Comments
by Peter Clarkson, Phil Gray, Joseph Fan, Jayne Godfrey,
Terry O’Keefe, Norman Wong, and seminar participants at
the University of Tasmania, and participants at the 2002
AAANZ Conference and the 2002 EAA Annual Congress
are greatly appreciated. I acknowledge the financial support
from the UQ Business School.
1. Introduction
Despite growing numbers of studies on the
association between institutional ownership and
earnings management (e.g. Bushee, 1998;
Cheng and Reitenga, 2000; Koh, 2003), the
association between institutional ownership and
income smoothing is rarely investigated
(Carlson and Bathala, 1997).
1
This study
examines the association between institutional
ownership and the likelihood of a portfolio firm
smoothing reported earnings through accruals
management. Income smoothing in this study
refers to earnings management toward an
earnings trend to reduce earnings variability
(e.g., Beidleman, 1973).
2
The practice of
income smoothing is well established (e.g.,
Beidleman, 1973; Godfrey and Jones, 1999;
Ronen and Sadan, 1981; Subramanyam, 1996)
and firms are facing increasing pressure to
report smooth earnings. For example, a
prominent FORTUNE 500 firm’s CEO claims
that, “[t]he No. 1 job of management is to
smooth out earnings” (Loomis, 1999).
Managers have incentives to smooth reported
earnings as it can affect earnings predictability,
perceived risk of the firm, managers’ personal
wealth and job security, among others. That is,
1 Bushee (1998) examines earnings management via
cutting research and development expenditure; Cheng
and Reitenga (2000) examine accruals management;
whilst Koh (2003) examines income increasing accruals
management. Carlson and Bathala (1997) is the only
known study directly investigate the association
between institutional ownership and income smoothing.
There are no known studies on the association between
institutional ownership and income smoothing in
Australia.
2 For this study, a firm is smoothing income if
Abs(Earnings-Trend)0.10).
Spearman correlations between the
explanatory variables are documented in
Table 3. Total institutional ownership (PIO) is
positively associated with firm size (SIZE)
consistent with the popularity of indexing
strategies among institutional investors in
Australia (Stapledon, 1996a; 1996b). The
positive association between institutional
ownership and firm size is also consistent
with the U.S. evidence (e.g., Carlson and
Bathala, 1997). Some weak evidence that
institutional ownership is negatively associated
with leverage also exists (p0.10; Column 3 of
13 Incidentally, all profit (loss) firms have positive
(negative) non-discretionary earnings.
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ACCOUNTING RESEARCH JOURNAL VOLUME 18 NO 2 (2005)
104
Table 4
Logit Regression of Income Smoother on Institutional Ownership Variables
and Control Variables
Prob(SMOOTH
= 1) = ? + ?
1
PIO + ?
2
SIZE + ?
3
LEV + ?
4
MGRSH + ?
5
CFO + ?
6
MINE +
!
=
1997
1994 k
?
k
YEAR
k
Estimated Coefficients (p-value)*
Variable Expected Sign
Full Sample Profit Firms
#
Loss Firms
##
(1) (2) (3)
Constant ? 2.1074
(0.0641)
2.1109
(0.1101)
2.1490
(0.5556)
PIO + 1.4505
(0.0283)
2.2064
(0.0068)
-2.1141
(0.1687)
SIZE +/- -0.1579
(0.1514)
-0.2445
(0.0638)
0.2513
(0.4838)
LEV - -0.8587
(0.0579)
-0.3280
(0.3409)
-1.2319
(0.1390)
MGRSH + 0.6729
(0.2290)
1.0316
(0.1620)
-3.3449
(0.1068)
CFO - -2.0895
(0.0795)
-1.0267
(0.3018)
-6.0148
(0.0620)
MINE +/- -0.2946
(0.4824)
-0.0504
(0.9227)
-0.7425
(0.6515)
Nagelkerke R
2
0.0980 0.1329 0.4734
Sample size (N) 202 156 46
* 1-tail p-value is reported if direction of the coefficient is predicted, otherwise 2-tail p-value is reported
# Profit firms all have NDE > 0
## Loss firms all have NDE < 0
PIO = Percentage institutional ownership
SIZE = Natural logarithm of total assets
LEV = Ratio of total liabilities to total tangible assets
MGRSH = Ratio of directors’ shareholdings (direct, indirect and beneficial holdings) to total ordinary shares outstanding
CFO = Cash flows from operations/total assets
MINE = Mining company dummy variables (1 if a firm is a mining firm; 0 otherwise)
SMOOTH = 1 if a firm an income smoother, i.e., Abs(Earnings – Trend) < Abs(NDE – Trend); 0 otherwise
Earnings = Current year earnings before interests and tax and before extraordinary items divided by prior year total assets
Trend = Prior year earnings before interests and tax and before extraordinary items divided by prior year total assets
NDE = Non-discretionary earnings as measured by earnings less discretionary accruals (DACC) as proxied by
modified Jones accruals model
Table 4). These findings are consistent with
existing argument that earnings management
incentives, including income smoothing
incentives, are likely to be stronger for profit
than loss firms, and hence their earnings
management strategies are likely to differ.
4.3 Are Income Smoothing Incentives
Uniform Across All Circumstances?
As posited in Section 2.2, it is also expected
that the predicted positive association between
institutional ownership and income smoothing
is more prevalent among firms with pre-
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
Institutional Ownership and Income Smoothing: Australian Evidence
105
Table 5
Logit Regression of Income Smoother on Institutional Ownership Variables
and Control Variables (Additional Tests)
Prob(SMOOTH
= 1) = ? + ?
1
PIO + ?
2
SIZE + ?
3
LEV + ?
4
MGRSH + ?
5
CFO + ?
6
MINE +
!
=
1997
1994 k
?
k
YEAR
k
Estimated Coefficients (p-value)*
Constant PIO SIZE LEV MGRSH CFO MINE Nagelkerke
? + +/- - + - +/- R
2
Panel A: Firms with NDE above their earnings trend (n=110)
1.2486
(0.4211)
3.0255
(0.0034)
-0.1732
(0.2902)
-0.0769
(0.4647)
2.0358
(0.0620)
0.7385
(0.3750)
-0.6730
(0.2460)
0.1810
Panel B: Firms with NDE below their earnings trend (n=92)
2.0665
(0.3091)
0.3867
(0.3850)
-0.0185
(0.9203)
-1.9516
(0.0426)
-1.2823
(0.2153)
-5.0078
(0.0251)
0.3016
(0.6977)
0.2384
Panel C: Profit firms with NDE above their earnings trend (n=102)
0.8587
(0.5935)
3.1538
(0.0036)
-0.1848
(0.2844)
0.2459
(0.4019)
2.4865
(0.0457)
1.2061
(0.3130)
-0.4577
(0.4973)
0.2170
Panel D: Profit firms with NDE below their earnings trend (n=54)
9.3702
(0.7124)
1.2587
(0.2585)
-0.3200
(0.2438)
-1.7281
(0.1763)
-1.6635
(0.1901)
-8.4417
(0.0386)
0.9215
(0.4035)
0.4053
Panel E: Firms with positive discretionary accruals (n=105)
0.7780
(0.6274)
2.0976
(0.0397)
-0.2741
(0.1223)
0.2021
(0.4161)
-0.5918
(0.3439)
-5.3352
(0.0092)
0.7224
(0.2477)
0.1900
*1-tail p-value is reported if direction of the coefficient is predicted, otherwise 2-tail p-value is reported
PIO = Percentage institutional ownership
SIZE = Natural logarithm of total assets
LEV = Ratio of total liabilities to total tangible assets
MGRSH = Ratio of directors’ shareholdings (direct, indirect and beneficial holdings) to total ordinary shares outstanding
CFO = Cash flows from operations/total assets
MINE = Mining company dummy variables (1 if a firm is a mining firm; 0 otherwise)
SMOOTH = 1 if a firm an income smoother, i.e., Abs(Earnings – Trend) < Abs(NDE – Trend); 0 otherwise
Earnings = Current year earnings before interests and tax and before extraordinary items divided by prior year total assets
Trend = Prior year earnings before interests and tax and before extraordinary items divided by prior year total assets
NDE = Non-discretionary earnings as measured by earnings less discretionary accruals (DACC) as proxied by
modified Jones accruals model
managed earnings above their earnings trend
(i.e., NDE > Trend).
14
To investigate this, the
logit regression is regressed on two sub-
samples, viz., firms with pre-managed earnings
(NDE) above versus below their earnings trend
14 To the extent that firms with NDE greater than their
earnings trend do not manage earnings downwards to
smooth earnings, it would bias against finding results
consistent with this expectation.
(results are reported in Panels A and B of Table
5 respectively).
15
For firms with NDE above
their earnings trend, the estimated coefficient of
the institutional ownership is positive and
significant (p=0.0034), while firms with NDE
15 Income smoothers among firms with pre-managed
earnings above (below) their earnings trend correspond
to firms that smooth their earnings down (up) towards
their earnings trend.
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
0
:
5
9
2
4
J
a
n
u
a
r
y
2
0
1
6
(
P
T
)
ACCOUNTING RESEARCH JOURNAL VOLUME 18 NO 2 (2005)
106
below their earnings trend have insignificant
coefficient (p=0.3850). These results suggest
that firms with high institutional ownership are
more likely to smooth income if they
are above their earnings trend before
accruals management. In contrast, institutional
ownership is not associated with income
smoothing when firms’ NDE is below their
earnings trend. Therefore, the initial findings
on Column 1 of Table 4 can be attributed
mainly to the association between institutional
ownership and income smoothing among firms
with pre-managed earnings above their
earnings trend.
The above findings also provide preliminary
empirical evidence indicating that the choices
between income smoothing and other earnings
management strategies that a manager faces
will depend upon whether a firm’s pre-
managed earnings is higher or lower than its
earnings trend. The results suggest that such
incentive trade-offs are less ambiguous among
firms with pre-managed earnings that exceed
their earnings trend than those below the trend.
Income Smoothing Incentives of Profit
versus Loss Firms
Further tests are conducted to examine whether
the differences between firms with NDE above
and below their earnings trend persist when
only profit firms are considered. Consistent
with the preceding findings, for profit firms
with NDE that exceed their earnings trend,
institutional ownership (PIO) is significantly
positively associated with income smoothing at
the 1% level (Table 5, Panel C). However, for
profit firms with NDE below their earnings
trend, their institutional ownership is not
associated with income smoothing (Table 5,
Panel D).
Combining the results from Panels B and D
of Table 5 indicates that for firms below their
earnings trend prior to accruals management,
institutional ownership is not associated with
income smoothing regardless of whether they
are profit or loss making firms.
16
16 For brevity, the results for loss firms with NDE below
or above their earnings trend are not reported. These
results reveal that institutional ownership is not
associated with income smoothing (p-values>0.10),
consistent with earlier findings.
Among the control variables, cash flows
from operations (CFO) are negatively
associated with income smoothing only among
firms with NDE below their earnings trend (p-
values