You are on page 1of 6

JID: RACREG

ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

Research in Accounting Regulation 0 0 0 (2017) 1–6

Contents lists available at ScienceDirect

Research in Accounting Regulation


journal homepage: www.elsevier.com/locate/racreg

Research report

Enhanced disclosure of other comprehensive income and increased


usefulness of net income: The implications of Accounting Standards
Update 2011–05
Linna Shi a,∗, Ping Wang b, Nan Zhou c
a
University of Cincinnati, USA
b
Pace University, USA
c
State University of New York at Binghamton, USA

a r t i c l e i n f o a b s t r a c t

Article history: Accounting Standards Update (ASU) 2011–05 eliminates the option to present other comprehensive in-
Available online xxx come (OCI) in the statement of changes in stockholders’ equity. This study empirically investigates
whether this mandatory change of OCI presentation format achieves FASB’s stated objective of improving
Keywords:
Other comprehensive income the transparency of financial reporting. First, ASU 2011–05 is found to greatly reduce the continuity of
Presentation format OCI from one period to the next. As OCI items are transitory in nature, the increased OCI volatility makes
ASU 2011–05 firms’ inherent risk more transparent to investors. Second, ASU 2011–05 is found to significantly increase
the ability of net income to influence stock prices. As OCI and net income are intertwined, the more
salient presentation of OCI enables investors to better interpret earnings. Supporting FASB’s position that
OCI items need to be more prominently displayed, these findings suggest that the new standard improves
transparency and usefulness of the reported OCI information.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction cial statement preparers, believed that the location of OCI dis-
closure could influence investors’ judgments (Yen, Hirst, & Hop-
Comprehensive income, the change of stockholders’ equity from kins, 2007). However, about 60% of the comment letters sent to
non-owner sources in a given period, consists of net income and the FASB reacted negatively to the proposal, stating that investors
other comprehensive income (OCI).1 How to present comprehen- would be unable to determine which measure, net income or com-
sive income has been at the center of regulatory attention for the prehensive income, was the appropriate one for investment deci-
past two decades. In June 1997, the Financial Accounting Standards sions (Du, McEnroe, & Stevens, 2016; FASB, 1997). Relenting to this
Board (FASB) issued Statement of Financial Accounting Standards pressure, the FASB in the final version of SFAS 130 allows a third,
(SFAS) 130 Reporting Comprehensive Income, which is now part of non-performance-based option, which is to present OCI in the rela-
Accounting Standards Codification (ASC) Topic 220 Comprehensive tively obscure statement of changes in stockholders’ equity (Jordan
Income. Consistent with the suggestion by the Association for In- & Clark, 2014).2
vestment Management and Research (AIMR, 1993) that a clear dis- The non-performance-based option has become the prevalent
play of OCI components in a performance statement would en- format to disclose OCI since the implementation of SFAS 130, be-
hance the transparency of financial reporting, the exposure draft cause managers believe that the greater volatility of firm perfor-
of SFAS 130 initially required that OCI and its components be pre- mance under the performance-based option could hurt the firm’s
sented in two performance-based alternatives, either in a single stock price and thus jeopardize their own careers (Bamber, Jiang,
continuous statement of income or in a separate statement of com- Petroni, & Wang, 2010; Graham, Harvey, & Rajgopal, 2005). In a
prehensive income (FASB, 1996). Many respondents, mostly finan- random sample of 100 NYSE firms, 89 reported OCI in a state-
ment of changes in stockholders’ equity, two in a continuous
statement with net income, and nine in a separate statement of

Corresponding author.
E-mail addresses: shiln@ucmail.uc.edu (L. Shi), pwang@pace.edu (P. Wang),
nzhou@binghamton.edu (N. Zhou). 2
On the contrary, IAS 1 Presentation of Financial Statements permits an entity to
1
OCI includes net unrealized holding gains and losses on some securities, gains present OCI either in a single statement of comprehensive income or in two sepa-
and losses from amendments to postretirement benefit plans, deferred gains and rate statements of net income and other comprehensive income (International Ac-
losses on derivatives, and adjustments from foreign currency translation. counting Standards Board (IASB), 2007).

https://doi.org/10.1016/j.racreg.2017.09.005
1052-0457/© 2017 Elsevier Ltd. All rights reserved.

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005
JID: RACREG
ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

2 L. Shi et al. / Research in Accounting Regulation 000 (2017) 1–6

comprehensive income (Pandit & Philips, 2004). Partly because of This paper has policy implications for standard setters regarding
the widespread avoidance of presenting OCI in performance-based the OCI presentation. When presented in a performance statement,
reports, the FASB in June 2011 issued Accounting Standards Up- the OCI information should be more helpful to financial statement
date (ASU) 2011–05 Comprehensive Income (Topic 220) Presentation users in their assessment of firms’ business activities and future
of Comprehensive Income, which eliminated the option of present- cash flows (FASB, 2011). Supporting FASB’s position, the findings
ing OCI in the statement of changes in stockholders’ equity and suggest that ASU 2011–05 improves transparency and usefulness
limits companies to the choice of either the income statement of the reported OCI information. Allowing a separate statement
or the statement of comprehensive income (Eaton, Easterday, & of comprehensive income, the FASB (2011, 45) “reasoned that this
Rhodes, 2013).3 two-statement approach achieves the same objectives as a single
Does the more prominent display of OCI items help achieve ASU statement because the statements would be consecutive and would
2011–05’s intended objective of improving the comparability, con- end with a total for comprehensive income, adequately increas-
sistency, and transparency of financial reporting (FASB, 2011)? Dif- ing the prominence of other comprehensive income.” Evidence that
ferent theories exist as to whether the presentation format matters OCI presented in a continuous income statement has the same ef-
for the transparency and valuation of accounting information. If in- fects on earnings continuity and value implications as OCI in a
vestors are rational in an efficient market, then the presentation statement of comprehensive income reinforces FASB’s notion that
format should not matter, as public information disclosed in any the key difference is between a performance report and a non-
format is fully incorporated into the stock price (Eaton et al., 2013). performance report, rather than between two different types of
In a departure from the efficient market hypothesis, Hirshleifer and performance reports.
Teoh (2003) assume that investors have limited attention and pro- This paper also contributes to two streams of the account-
cessing power when modeling firms’ choices between alternative ing literature. First, the study adds to the literature on finan-
means of presenting information. Due to limited attention, in- cial statement presentation by documenting that the more promi-
vestors can process and absorb salient information more easily nent display of OCI in a performance report improves the trans-
than less salient information implicit in the public domain. They parency of financial reporting. Previous studies, only focusing on
further conclude that disclosing equivalent information in different items within the income statement, find that the closer the line
formats can have different effects on investors’ perceptions due to item to the top line in the income statement, the greater its ef-
investors’ limited attention. Drawing on this limited attention the- fect on stock prices or future earnings (Bartov & Mohanram, 2014;
ory (Hirshleifer & Teoh, 2003), the current study hypothesizes that Bradshaw & Sloan, 2002; Fairfield, Sweeney, & Yohn, 1996; Lipe,
the new presentation requirement in ASU 2011–05 makes financial 1986; Ohlson & Penman, 1992; Strong & Walker, 1993). Second,
statements more transparent and thus more useful to investors. the study adds to the OCI literature that has yielded mixed ev-
The sample consists of 1663 firm-year observations pertaining idence so far. While experimental studies find that OCI informa-
to S&P 500 firms from 2010 to 2013, two years before and two tion reported in the income statement can help investors extract
years after the ASU 2011–05 implementation. Two key findings are useful information (Hirst & Hopkins, 1998; Maines & McDaniel,
as follows. First, ASU 2011–05 reduces the continuity of net income 20 0 0), archival studies show that investors only react to OCI infor-
from one year to the next by 7% and accelerates the reversal of mation reported in the statement of changes in stockholders’ eq-
OCI by 60%.4 Consistent with the notion that OCI items are transi- uity (Chambers, Linsmeier, Shakespeare, & Sougiannis, 2007; Lin,
tory in nature (Bamber et al., 2010; Barker, 2004; Linsmeier et al., Martinez, Wang, & Yang, 2017). Deviating from the assumption in
1997), the increased OCI volatility makes firms’ inherent risk more Chambers et al. (2007) and Lin et al. (2017) that the OCI presen-
transparent to investors (Huang, Lin, & Raghunandan, 2016). Sec- tation format affects only OCI, this paper conjectures that the OCI
ond, ASU 2011–05 significantly increases the ability of net income presentation format can also have an effect on net income. Specif-
to influence stock prices. Given that OCI and net income are in- ically, it is uncovered that displaying OCI in a performance report
tertwined, the more salient presentation of OCI enables investors enhances the ability of net income, rather than OCI, to influence
to better interpret earnings.5 After ASU 2011–05, more than 80% stock prices. While Collins, Maydew, and Weiss (1997) document
of S&P 500 firms choose to report OCI in a separate statement of the decline of earnings’ effect on stock prices in past decades, the
comprehensive income instead of a continuous income statement current study suggests that improving the salience of OCI would
(Kim, 2016). This calls for further examination of the pros and cons mitigate this trend by improving the ability of earnings to influ-
of these two types of performance reports. The supplemental anal- ence stock prices. These findings, new to the OCI literature, provide
ysis for the post ASU 2011–05 period indicates that the choice of evidence to the FASB that the more salient OCI presentation under
reporting OCI in either a continuous income statement or a state- ASU 2011–05 improves the usefulness of earnings.
ment of comprehensive income does not have any differential in- The rest of the paper is structured as follows. The next section
fluence on the continuity of earnings from one period to the next discusses sample selection and descriptive statistics. The third sec-
or the ability of earnings to influence stock prices. tion presents empirical findings. The last section concludes.

3
2. Sample selection and descriptive statistics
Effective for fiscal years beginning after December 15, 2011, ASU 2011–05 fo-
cuses on the change of the OCI presentation format. Other issues, such as the calcu-
lation of OCI items, the reporting about the reclassification from OCI to net income, Table 1 describes the sample selection. Starting with the Com-
and the option to report OCI either before tax or net of tax, are not changed. The pustat S&P 500 Index constituents in 2010, the initial sample tracks
FASB made the change to increase the prominence of OCI information, which may these 500 firms over from 2010 to 2013, two years before and two
improve investors’ recognition and understanding of OCI. In addition, the change
years after ASU 2011–05.6 The OCI presentation format informa-
also makes U.S. GAAP and international financial reporting standards (IFRS) more
convergent in terms of OCI presentation (FASB, 2011). tion is hand-collected from firms’ 10-K filings. The sample loses 36
4
Similarly, Jones and Smith (2011) find that OCI gains and losses exhibit rever- observations due to missing 10-K filings7 ; 38 observations due to
sal, which may be due to the reclassification of gains and losses out of the balance
sheet account Accumulated Other Comprehensive Income (AOCI) and into net in-
6
come. They illustrate this scenario with an example in footnote 11 on page 2052. This sample construction is driven by the adoption of a pre-post quasi-
5 experiment research design, where the treatment effect is ASU 2011-05 issued by
Note that OCI itself does not contribute to a firm’s value, which is consistent
with the evidence that OCI adds little information to net income in explaining stock the FASB.
7
prices (Dhaliwal, Subramanyam, & Trezevant, 1999; O’Hanlon & Pope, 1999; Cahan, One or multiple years of 10-K filings are missing for 19 firms, which translates
Courtenay, Gronewaller, & Upton, 20 0 0). to 36 missing firm-year observations. Specifically, the 10-K filings are missing for 17

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005
JID: RACREG
ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

L. Shi et al. / Research in Accounting Regulation 000 (2017) 1–6 3

Table 1 (coefficient on C ItA + that on C ItA ∗ DY ) after ASU 2011–05. The third
Sample selection.
column compares earnings continuity when firms report OCI in a
S&P 500 firms in 2010 tracked for a four year period from 2010 to 2013 20 0 0 performance statement (i.e., IS or SCI) versus a non-performance
Less: statement (i.e., SCSE). The percentage is lowered from 87% (coef-
Firm-years with missing 10-K filings (36)
ficient on CItA ) with OCI presented in SCSE to 72% (coefficient on
Firm-years with missing OCI in the 10-K filings (38)
Firm-years with missing beginning market value or total assets (190) CItA + that on CItA ∗ DIS&SCI ) with OCI in either IS or SCI.
Firm-years with missing stock returns for the current year (28) Panel B decomposes CItA into net income (NItA ) and other com-
Firm-years with missing comprehensive income for the next year (35) prehensive income (OCItA ). The first column shows that 87% (co-
Final sample 1663 efficient on NItA ) of current net income continues to the next pe-
This table describes the sample selection process. riod, whereas 58% (the coefficient on OCItA ) of current OCI expe-
riences a reversal in the next period. Jones and Smith (2011) find
missing OCI information in a financial statement8 ; and 253 obser- that OCI gains and losses exhibit similar reversal, which is likely
vations due to missing variables in Compustat and CRSP. The final caused by the reclassification of gains and losses out of AOCI and
sample consists of 1663 firm-year observations. Table 2 reports the into net income. The second column shows that ASU 2011–05 re-
distribution based on firms’ OCI presentation formats. Before ASU duces the continuity of net income by 7% (coefficient on NItA ∗ DY )
2011–05, the percentage of firms reporting OCI in the statement of and increase the reversal of OCI by 60% (coefficient on OCItA ∗ DY ).
changes in stockholders’ equity (SCSE hereafter) is 74.58% (63.51%) Similarly, the third column documents that the continuity of net
in 2010 (2011). After ASU 2011–05, this percentage drops to 9.72% income falls by 10% (coefficient on NItA ∗ DIS&SCI ) and the reversal
(0.76%) in 2012 (2013). Out of 396 firms in 2013, two years after of OCI increases by 72% (coefficient on OCItA ∗ DIS&SCI ) when OCI is
the adoption of ASU 2011–05, 372 firms (93.94%) choose the state- presented in a performance statement. Consistent with the notion
ment of comprehensive income (SCI hereafter) to report OCI, 21 that OCI items are transitory in nature (Bamber et al., 2010; Barker,
firms (5.30%) adopt the continuous statement of income (IS here- 2004; Linsmeier et al., 1997), the increased OCI volatility post ASU
after), and three firms (0.76%) continue to use the disallowed SCSE. 2011–05 makes firms’ inherent risk more transparent to investors
Table 3 tabulates the descriptive statistics for regression variables.9 (Huang et al., 2016).
Comprehensive income (CI), net income (NI) and OCI values are
hand-collected from sample firms’ 10-K filings.10 The mean CI, NI 3.2. Ability to influence stock prices
and OCI, divided by average total assets (beginning market value),
are 7.49% (6.85%), 7.46% (6.86%), and −0.11% (−0.03%), respectively. Table 5 provides evidence on whether ASU 2011–05 signifi-
The average annual stock return is 19.13% for the sample period cantly increases the ability of earnings to influence stock prices
from 2010 to 2013. by regressing current year’s stock returns (RETt ) on current year’s
comprehensive income (CItMV ) or its two components.12 RETt is the
3. Empirical findings 12-month raw cumulative stock returns starting from the fifth
month of the last fiscal year-end to the fourth month after the cur-
3.1. Continuity of earnings from one period to the next rent fiscal year-end. DY and DIS&SCI are as defined in Table 4. Panel
A focuses on total comprehensive income. The first column pro-
Table 4 presents findings on whether ASU 2011–05 reduces the vides a baseline model regarding the effect of total comprehensive
continuity of comprehensive income by regressing next year’s to- income on stock returns. The significant coefficient on CItMV shows
tal comprehensive income (CIt+1A ) on either current year’s com-
that stock returns are contemporaneously associated with total
A
prehensive income (CIt ) or its two components.11 For the sample comprehensive income. The second column finds that the effect of
period, DY is an indicator variable equal to one if a firm’s fiscal total comprehensive income on investors’ valuation increases from
year ends in 2012 and 2013, and zero in 2010 and 2011. DIS&SCI 0.35 (coefficient on CItMV ) before ASU 2011–05 to 0.88 (coefficient
is an indicator variable equal to one if a firm reports OCI in IS on CItMV +that on CItMV ∗ DY ) after ASU 2011–05. The third column
or SCI, and zero if a firm presents OCI in SCSE. Panel A focuses investigates whether the value implications of comprehensive in-
on total comprehensive income. The first column provides a base- come are different when OCI is presented in IS or SCI as opposed
line model for the continuity of comprehensive income. The co- to SCSE. The effect of total comprehensive income on stock returns
efficient on CItA indicates that 78% of total comprehensive income is 0.33 (coefficient on CItMV ) when OCI is presented in SCSE, versus
continues from the current period to the next over the entire sam- 0.76 when OCI is presented in IS or SCI (coefficient on CItMV + that
ple period. The second column examines whether earnings con- on CItMV ∗ DIS&SCI ). These results indicate that investors react to to-
tinuity changes after ASU 2011–05. The percentage drops signifi- tal comprehensive income more strongly when OCI is reported in
cantly from 85% (coefficient on CItA ) before ASU 2011–05 to 72% a performance report post ASU 2011–05.
Panel B decomposes CItMV into net income (NItMV ) and other
comprehensive income (OCItMV ). The first column shows that the
firms in one or multiple years between 2010 and 2012, but are available in 2013.
pricing on OCI is almost one to one (coefficient on OCItMV = 1.04),
The 10-K filings are missing for one firm over the entire sample period from 2010 to
2013. One firm, Life Technologies Corp., has a 10-K filing in 2010, 2011 and 2012, but
while the pricing on net income is lower than one (coefficient on
not in 2013. Since survivorship claims only one missing firm-year observation (Life NItMV = 0.70). The second column reports that ASU 2011–05 sub-
Technologies Corp. in 2013), the results in this paper are unlikely to be influenced stantially increases the ability of net income to affect stock returns
by the survivorship bias. (significant coefficient on NItMV ∗ DY = 0.86) but does not change
8
The OCI information either cannot be found in the 10-K filings or is disclosed
the ability of OCI to influence stock returns (insignificant coeffi-
in a footnote rather than in the main body of a financial statement.
9
For all regression variables, top 1% extreme values on the left (right) tail are
cient on OCItMV ∗ DY ). The third column finds that the effect of net
replaced by the value at the first (99th ) percentile, so as to mitigate the influence income on stock returns is higher when OCI is presented in either
of outliers in regressions. IS or SCI (significant coefficient on NItMV ∗ DIS&SCI = 0.74), whereas
10
There could be a discrepancy between CI and NI from 10-K filings and those
from Compustat, due to the fact that Compustat attributes CI and NI sometimes to
12
controlling holders or sometimes to the entire company. This paper uses CI, NI and The superscript MV denotes that the variable is scaled by beginning mar-
OCI values for the entire company, without considering non-controlling (minority) ket value, which is equal to the stock price at the end of the last fiscal period
shareholders. end (PRCC_F) multiplied by the total number of shares outstanding (CSHO) (e.g.,
11
The superscript A denotes that the variable is scaled by average total assets. Chambers et al., 2007).

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005
JID: RACREG
ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

4 L. Shi et al. / Research in Accounting Regulation 000 (2017) 1–6

Table 2
Sample distribution.

Year OCI presentation formats

Statement of changes in stockholders’ equity Income statement Statement of comprehensive income Total
(SCSE) (IS) (SCI)

2010 308 11 94 413


74.58% 2.66% 22.76%
2011 268 17 137 422
63.51% 4.03% 32.46%
2012 42 28 362 432
9.72% 6.48% 83.80%
2013 3 21 372 396
0.76% 5.30% 93.94%
Total 621 77 965 1663
37.34% 4.63% 58.03%

This table shows the annual distribution pertaining to different presentation formats of other comprehensive income (OCI).
The OCI presentation format information is hand-collected by reading sample firms’ 10-K filings from the EDGAR Data by the
Securities and Exchange Commission.
Table 3 Table 4
Descriptive statistics. Earnings continuity for different OCI presentation formats.

Variable Mean Q1 Median Q3 Std Dev N Panel A: Total comprehensive income (Dependent variable = CIt+1
A
)

A
CIt+1 0.0773 0.0724 0.0286 0.0632 0.1118 1663 Variable Model 1.1 Model 1.2 Model 1.3
CItA 0.0749 0.0633 0.0321 0.0639 0.1074 1663 CItA 0.78 0.85 0.87
NItA 0.0746 0.0610 0.0322 0.0644 0.1072 1663
OCItA −0.0011 0.0148 −0.0056 -0.0 0 01 0.0033 1663 (0.00) (0.00) (0.00)
RETt 0.1913 0.2604 0.0216 0.1782 0.3329 1663 CItA ∗ DY −0.13
CItMV 0.0685 0.0561 0.0453 0.0660 0.0873 1663 (0.00)
NItMV 0.0686 0.0448 0.0499 0.0676 0.0858 1663 DY 0.01
OCItMV −0.0 0 03 0.0254 −0.0053 −0.0 0 01 0.0035 1663 (0.06)
DY 0.4979 0.5001 0.0 0 0 0 0.0 0 0 0 1.0 0 0 0 1663 CItA ∗ DIS&SCI −0.15
DIS&SCI 0.6266 0.4839 0.0 0 0 0 1.0 0 0 0 1.0 0 0 0 1663 (0.00)
DIS 0.0463 0.2102 0.0 0 0 0 0.0 0 0 0 0.0 0 0 0 783 DIS&SCI 0.00
(0.38)
This table presents the descriptive statistics for regression variables. CI, NI and Intercept 0.02 0.01 0.02
OCI are, respectively, total comprehensive income, net income and other com- (0.00) (0.00) (0.00)
prehensive income. The superscript A (MV ) denotes that the variable is scaled by Number of Obs. 1663 1663 1663
average total assets (beginning market value). The subscript t (t+1 ) denotes that Adjusted R2 46.34% 46.64% 46.98%
the variable is in the current year (the next year). RETt is the 12-month raw cu- F-value 1436.33 485.17 491.93
mulative stock returns starting from the fifth month of the last fiscal year-end Prob > F 0.00 0.00 0.00
to the fourth month after the current fiscal year-end. DY is an indicator variable
Panel B: Decomposition of comprehensive income (Dependent variable = CIt+1
A
)
equal to one if a firm’s fiscal period ends in 2012 and 2013, and zero if it ends in
2010 and 2011. DIS&SCI is an indicator variable equal to one if a firm reports OCI Variable Model 2.1 Model 2.2 Model 2.3
in either the income statement (IS) or the statement of comprehensive income
(SCI), and zero if a firm presents OCI in the statement of changes in stockhold- NItA 0.87 0.91 0.93
ers’ equity (SCSE). For the subsample of post ASU 2011–05 observations, DIS is an (0.00) (0.00) (0.00)
indicator variable equal to one if a firm presents OCI in the income statement, OCItA −0.58 −0.30 −0.14
and zero if a firm presents OCI in the statement of comprehensive income. (0.00) (0.01) (0.28)
NItA ∗ DY −0.07
(0.06)
OCItA ∗ DY −0.60
the effect of OCI on stock returns is not affected by the OCI pre-
(0.00)
sentation format (insignificant coefficient on OCItMV ∗ DIS&SCI ). Prior DY 0.01
studies (e.g., Chambers et al., 2007; Lin et al., 2017) find that the (0.01)
value implications of OCI are not affected by the OCI presenta- NItA ∗ DIS&SCI −0.10
tion format and thus question whether ASU 2011–05 improves in- (0.02)
OCItA ∗ DIS&SCI −0.72
vestors’ understanding of earnings information. Unlike these stud-
(0.00)
ies, which implicitly assume that the effect of net income on stock DIS&SCI 0.01
returns is independent of the OCI reporting location, this paper (0.12)
provides evidence that the more salient OCI under ASU 2011–05 Intercept 0.01 0.01 0.01
(0.00) (0.01) (0.01)
helps investors better understand net income and thus increases
net income’s ability to influence stock prices. Number of Obs. 1663 1663 1663
Adjusted R2 54.64% 55.17% 55.24%
F-value 1002.18 410.12 411.16
3.3. Further analysis Prob > F 0.00 0.00 0.00

This table presents the regression results on the continuity of comprehensive in-
ASU 2011–05 allows two types of performance reports (i.e., IS
come (CI), net income (NI), and other comprehensive income (OCI), based on firms’
and SCI). Since Table 2 reports that over 90% of the sample firms OCI presentation formats. The superscript A denotes that the variable is scaled by
favor SCI, Table 6 further studies whether SCI and IS have dif- average total assets. The subscript t (t+1 ) denotes that the variable is in the current
ferent effects on earnings continuity and value implications. The year (the next year). DY is an indicator variable equal to one if a firm’s fiscal pe-
analysis is restricted to a subsample of 783 observations that re- riod ends in 2012 and 2013, and zero if in 2010 and 2011. DIS&SCI is an indicator
variable equal to one if a firm reports OCI in either the income statement (IS) or
port OCI in a performance report for the post ASU 2011–05 pe- the statement of comprehensive income (SCI), and zero if a firm presents OCI in
riod of 2012 and 2013. DIS is a dummy variable equal to one if the statement of changes in stockholders’ equity (SCSE). For each explanatory vari-
a firm discloses its OCI information in IS, and zero if OCI is re- able, the coefficient estimate is reported on the top, and the p-value based on a
ported in SCI. All other variables are as defined in Tables 4 and 5. two-tailed t-test is reported in parentheses at the bottom.

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005
JID: RACREG
ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

L. Shi et al. / Research in Accounting Regulation 000 (2017) 1–6 5

Table 5 Table 6
Value implications for different OCI presentation formats. OCI presentation formats under ASU 2011–05: Income statement versus statement
of comprehensive income.
Panel A: Total comprehensive income (Dependent variable = RETt )
Earnings continuity Value implications
Variable Model 3.1 Model 3.2 Model 3.3
Dependent variable = CIt+1
A
Dependent variable = RETt
CItMV 0.63 0.35 0.33
Variable Model 1.4 Model 2.4 Variable Model 3.4 Model 4.4
(0.00) (0.03) (0.10)
CItMV ∗ DY 0.53 CItA 0.70 CItMV 0.79
(0.02) (0.00) (0.00)
DY 0.05 NItA 0.84 NItMV 1.10
(0.01) (0.00) (0.00)
CItMV ∗ DIS&SCI 0.43 OCItA −0.93 OCItMV 0.61
(0.08) (0.00) (0.07)
DIS&SCI 0.01 CItA ∗ DIS 0.07 CItMV ∗ DIS 1.03
(0.70) (0.50) (0.10)
Intercept 0.15 0.12 0.15 NItA ∗ DIS 0.03 NItMV ∗ DIS 1.15
(0.00) (0.00) (0.00) (0.72) (0.15)
OCItA ∗ DIS −0.51 OCItMV ∗ DIS 0.36
Number of Obs. 1663 1663 1663
(0.29) (0.81)
Adjusted R2 1.76% 4.73% 2.29%
DIS 0.01 0.00 DIS −0.07 −0.07
F-value 30.78 28.53 14.00
(0.68) (0.73) (0.16) (0.21)
Prob > F 0.00 0.00 0.00
Intercept 0.02 0.02 Intercept 0.19 0.16
Panel B: Decomposition of comprehensive income (Dependent variable = RETt ) (0.00) (0.00) (0.00) (0.00)
Number of Obs. 783 783 Number of Obs. 783 783
Variable Model 4.1 Model 4.2 Model 4.3 2 2
Adjusted R 40.77% 53.47% Adjusted R 3.98% 5.07%
NItMV 0.70 0.31 0.23 F-value 180.39 180.72 Number of Obs. 11.81 9.35
(0.00) (0.12) (0.35) Prob > F 0.00 0.00 Prob > F 0.00 0.00
OCItMV 1.04 1.01 1.40
This table presents the regression results on the earnings continuity and value im-
(0.00) (0.01) (0.00)
plications of comprehensive income (CI), net income (NI), and other comprehensive
NItMV ∗ DY 0.86
income (OCI), depending on a firm’s choice of presenting OCI either in an income
(0.00)
statement (IS) or in a statement of comprehensive income (SCI). This analysis is
OCI/MVt ∗ DY −0.27
restricted to a subsample of 783 observations for the post ASU 2011–05 period of
(0.59)
2012 and 2013. The superscript A (MV ) denotes that the variable is scaled by aver-
DY 0.03
age total assets (or beginning market value). The subscript t (t+1 ) denotes that the
(0.26)
variable is in the current year (the next year). RETt is the 12-month raw cumulative
NI/MVt ∗ DIS&SCI 0.74
stock returns starting from the fifth month of the last fiscal year-end to the fourth
(0.01)
month after the current fiscal year-end. DIS is an indicator variable equal to one if a
OCI/MVt ∗ DIS&SCI −0.68
firm presents OCI in IS, and zero if a firm presents OCI in SCI. For each explanatory
(0.22)
variable, the coefficient estimate is reported on the top, and the p-value based on
DIS&SCI −0.02
a two-tailed t-test is reported in parentheses at the bottom.
(0.45)
Intercept 0.14 0.13 0.16
(0.00) (0.00) (0.00)
Number of Obs. 1663 1663 1663 to present OCI under ASU 2011–05 does not affect earnings conti-
Adjusted R2 2.47% 5.49% 3.15% nuity or value implications. This reinforces FASB’s notion that the
F-value 22.05 20.30 11.81
Prob > F 0.00 0.00 0.00
key difference in OCI disclosure is between a performance report
and a non-performance report, rather than between two different
This table presents the regression results on the value implications of comprehen- types of performance reports.
sive income (CI), net income (NI), and other comprehensive income (OCI), based
on firms’ OCI presentation formats. The superscript MV denotes that the variable is
scaled by beginning market value. The subscript t denotes that the variable is in the
current year. RETt is the 12-month raw cumulative stock returns starting from the 4. Conclusion
fifth month of the last fiscal year-end to the fourth month after the current fiscal
year-end. DY is an indicator variable equal to one if a firm’s fiscal period ends in
2012 and 2013, and zero if in 2010 and 2011. DIS&SCI is an indicator variable equal
This paper empirically investigates whether the mandatory
to one if a firm reports OCI in either the income statement (IS) or the statement change of OCI presentation format in ASU 2011–05 achieves the
of comprehensive income (SCI), and zero if a firm presents OCI in the statement FASB’s stated objective of improving the transparency of financial
of changes in stockholders’ equity (SCSE). For each explanatory variable, the coeffi- reporting. Main findings of this paper are as follows. First, ASU
cient estimate is reported on the top, and the p-value based on a two-tailed t-test
2011–05 is found to greatly reduce the continuity of OCI and net
is reported in parentheses at the bottom.
income from one period to the next. As OCI items are transitory
in nature, the increased OCI volatility makes firms’ inherent risk
more transparent to investors. Second, ASU 2011–05 is found to
The first two columns find that OCI presentation in IS or SCI does significantly increase the ability of net income to influence stock
not make any difference in the continuity of total comprehensive prices. As OCI and net income are intertwined, the more salient
income, net income and OCI (insignificant coefficients on CItA ∗ DIS , presentation of OCI enables investors to better interpret earnings.
NItA ∗ DIS and OCItA ∗ DIS ). The next two columns investigate whether Collectively, these findings suggest that the new standard improves
reporting OCI in IS versus SCI affects the value implications of transparency and usefulness of the reported OCI information. Nev-
earnings. The third column finds that when OCI is reported in IS, ertheless, caution needs to be taken in drawing inferences from
annual stock returns have a greater association with total com- the results, as this study is limited by the scope of its investiga-
prehensive income (coefficient on CItMV ∗ DIS significant at the 10% tion. The reliance on large S&P 500 firms, albeit consistent with
level). After decomposing comprehensive income into net income Kim (2016) and Lin et al. (2017), implies that the message cannot
and OCI, the fourth column shows that reporting OCI in either IS be generalized to medium and small public firms. Moreover, the
or SCI no longer makes any difference in the ability of net in- focus on the short-term effects of ASU 2011–05 indicates that the
come or OCI to influence stock returns (insignificant coefficients conclusion cannot be extended to the long-term consequences of
on NItMV ∗ DIS and OCItMV ∗ DIS ). Overall, whether IS or SCI is used ASU 2011–05.

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005
JID: RACREG
ARTICLE IN PRESS [m5G;November 8, 2017;9:1]

6 L. Shi et al. / Research in Accounting Regulation 000 (2017) 1–6

This paper has policy implications regarding current and future Eaton, T., Easterday, K., & Rhodes, M. (2013). The presentation of other comprehen-
FASB projects. In May 2014, the FASB launched the Simplification sive income: FASB’s recent and proposed changes. The CPA Journal, 83(3), 32–35.
Fairfield, P., Sweeney, R., & Yohn, T. (1996). Accounting classification and the predic-
Initiative to “improve or maintain the usefulness of the informa- tive content of earnings. The Accounting Review, 71(3), 337–355.
tion reported to investors while reducing costs and complexity in Financial Accounting Standards Board (FASB). (1996). Exposure draft: reporting com-
financial reporting”.13 As of July 2016, the FASB has nine completed prehensive income. Norwalk, CT: FASB.
Financial Accounting Standards Board (FASB). (1997). Statement of financial account-
and six ongoing simplification projects on a wide range of top- ing standards no. 130: reporting comprehensive income. Norwalk, CT: FASB.
ics (Jones, 2016). This initiative results in a sequence of Account- Financial Accounting Standards Board (FASB). (2011). Accounting standards update no.
ing Standards Updates issued or proposed by the FASB. Examples 2011-05: comprehensive income (Topic 220) presentation of comprehensive income.
Norwalk, CT: FASB.
include ASU 2015–03 Simplifying the Presentation of Debt Issuance
Financial Accounting Standards Board (FASB). (2015). Accounting standards update
Costs (FASB, 2015) and Proposed ASU Simplifying the Classification no. 2015-03: interest—imputation of interest (Subtopic 835-30) simplifying the pre-
of Debt in a Classified Balance Sheet (FASB, 2017). Since the stated sentation of debt issuance costs. Norwalk, CT: FASB.
Financial Accounting Standards Board (FASB). (2017). Proposed accounting standards
objective for the Simplification Initiative is similar to that for ASU
update (exposure draft): debt (Topic 470) simplifying the classification of debt
2011–05, the finding that the presentation format affects the trans- in a classified balance sheet (current versus noncurrent). Norwalk, CT: FASB.
parency and usefulness of financial information renders support to Graham, J., Harvey, C., & Rajgopal, S. (2005). The economic implications of corporate
the rationale behind FASB’s Simplification Initiative. financial reporting. Journal of Accounting and Economics, 40(1–3), 3–73.
Hirshleifer, D., & Teoh, S. H. (2003). Limited attention, information disclosure, and
financial reporting. Journal of Accounting and Economics, 36(1–3), 337–386.
Acknowledgements Hirst, D., & Hopkins, P. (1998). Comprehensive income reporting and analysts’ valu-
ation judgments. Journal of Accounting Research, 36(Supplement), 47–74.
Huang, H., Lin, S., & Raghunandan, K. (2016). The volatility of other comprehensive
We thank Gary Previts (Editor), two anonymous reviewers, Yang income and audit fees. Accounting Horizons, 30(2), 195–210.
Cheng, Jung Yeun Kim, and workshop participants at the 2017 International Accounting Standards Board (IASB). (2007). International accounting
American Accounting Association (AAA) Ohio Region Meeting in standards 1 presentation of financial statements. London, UK: IASB.
Jones, D., & Smith, K. (2011). Comparing the value relevance, predictive value, and
Columbus, Ohio and the 2017 AAA Annual Meeting in San Diego, persistence of other comprehensive income and special items. The Accounting
California for their helpful comments. Review, 86(6), 2047–2073.
Jones, J. (2016). The state of FASB’s simplification Initiative: Toward a simpler GAAP.
References The CPA Journal, 86(7), 56–59.
Jordan, C., & Clark, S. (2014). Reporting preferences under the comprehensive in-
come standard: Examining its use in practice. The CPA Journal, 84(5), 34–39.
Association for Investment Management and Research (AIMR). (1993). Financial re-
Kim, J. H. (2016). Presentation formats of other comprehensive income after Ac-
porting in the 1990 s and beyond. Charlottesville, VA: AIMR.
counting Standards Update 2011–05. Research in Accounting Regulation, 28(2),
Bamber, L., Jiang, J., Petroni, K., & Wang, I. (2010). Comprehensive income: Who’s
118–122.
afraid of performance reporting? The Accounting Review, 85(1), 97–126.
Lin, S., Martinez, D., Wang, C., & Yang, Y. (2017). Is other comprehensive income
Barker, R. (2004). Reporting financial performance. Accounting Horizons, 18(2),
reported in the income statement more value relevant? The role of financial
157–172.
statement presentation. Journal of Accounting, Auditing and Finance forthcoming.
Bartov, E., & Mohanram, P. (2014). Does income statement placement matter to in-
Linsmeier, T., Gribble, J., Jennings, R., Lang, M., Penman, S., Petroni, K., et al. (1997).
vestors? The case of gains/losses from early debt extinguishment. The Account-
An issues paper on comprehensive income. Accounting Horizons, 11(2), 120–126.
ing Review, 89(6), 2021–2055.
Lipe, R. (1986). The information contained in the components of earnings. Journal of
Bradshaw, M., & Sloan, R. (2002). GAAP versus the street: An empirical assessment
Accounting Research, 24(1), 37–64.
of two alternative definitions of earnings. Journal of Accounting Research, 40(1),
Maines, L., & McDaniel, L. (20 0 0). Effects of comprehensive-income characteristics
41–66.
on nonprofessional investors’ judgments: The role of financial statement pre-
Cahan, S., Courtenay, S., Gronewaller, P., & Upton, D. (20 0 0). Value relevance of man-
sentation format. The Accounting Review, 75(2), 177–204.
dated comprehensive income disclosures. Journal of Business Finance and Ac-
O’Hanlon, J., & Pope, P. (1999). The value relevance of UK dirty surplus accounting
counting, 27(9–10), 1273–1301.
flows. British Accounting Review, 31(4), 459–482.
Chambers, D., Linsmeier, T. J., Shakespeare, C., & Sougiannis, T. (2007). An evaluation
Ohlson, J., & Penman, S. (1992). Disaggregated accounting data as explanatory vari-
of SFAS No. 130 comprehensive income disclosures. Review of Accounting Studies,
ables for returns. Journal of Accounting, Auditing and Finance, 7(4), 553–573.
12(4), 557–593.
Pandit, G., & Phillips, J. (2004). Comprehensive income: Reporting preferences of
Collins, D. W., Maydew, E. L., & Weiss, I. S. (1997). Changes in the value-relevance
public companies. The CPA Journal, 74(11), 40–41.
of earnings and book values over the past forty years. Journal of Accounting and
Strong, N., & Walker, M. (1993). The explanatory power of earnings for stock returns.
Economics, 24(1), 39–67.
The Accounting Review, 68(2), 385–399.
Dhaliwal, D., Subramanyam, K., & Trezevant, R. (1999). Is comprehensive income
Yen, A., Hirst, D., & Hopkins, P. (2007). A content analysis of the comprehensive
superior to net income as a measure of firm performance. Journal of Accounting
income exposure draft comment letters. Research in Accounting Regulation, 19,
and Economics, 26(1–3), 43–67.
53–79.
Du, N., McEnroe, J., & Stevens, K. (2016). Why isn’t comprehensive income compre-
hensible. Strategic Finance, 98(5), 46–53.

13
Please refer to the FASB news release on June 10, 2014. http://www.fasb.org/cs/
ContentServer?pagename=FASB%2FFASBContent_C%2FNewsPage&cid=1176164118378

Please cite this article as: L. Shi et al., Enhanced disclosure of other comprehensive income and increased useful-
ness of net income: The implications of Accounting Standards Update 2011–05, Research in Accounting Regulation (2017),
https://doi.org/10.1016/j.racreg.2017.09.005