Early evidence of the volatility of comprehensive income and its components.

McCoy, Timothy L. ; Thompson, James H. ; Hoskins, Margaret A. 等

INTRODUCTION

The Financial Accounting Standards Board issued Statement of Financial Accounting Standard (SFAS) No. 130 Reporting Comprehensive Income, in June 1997, effective for fiscal periods beginning after December 15, 1997. Comprehensive Income is defined by FASB in SFAC No. 6 as the change in a firm's net assets (assets minus liabilities) from non-owner sources. Thus SFAS No. 130 is consistent with the Asset-Liability approach to income measurement whereby an increase in the value of net assets creates income, with comprehensive income capturing the overall increase or decrease in net assets for the period. Comprehensive income (CI) is comprised of net income and other comprehensive income (OCI). Other comprehensive income items consist primarily of gains and losses which by-pass the income statement under current GAAP and are carried straight to the owner's equity section of the balance sheet.

The major objective of SFAS No. 130 was to display Other Comprehensive Income items in a financial statement having equal prominence with other financial statements. While SFAS No. 130 requires that comprehensive income and its components be disclosed, it does not prescribe the specific method of disclosure. It does however suggest three alternatives: 1) a combined statement of net income and comprehensive income 2) a separate statement of comprehensive income and 3) within a statement of changes in equity. FASB encouraged the use of one of the first two methods.

BACKGROUND FOR ISSUANCE OF SFAS NO. 130

Historically, income presentation issues were primarily characterized in terms of a contrast between the current operating performance (dirty surplus) and the all-inclusive (clean surplus) approaches. Under the current operating performance concept of income, only ordinary and recurring revenues, expenses, gains, and losses are recognized as income while extraordinary and non-recurring gains and losses are excluded from income. Under the all-inclusive concept of income, however, all revenues, expenses, gains, and losses recognized during the period are included in income, regardless of whether they are considered to be results of normal, recurring operations of the period. The Accounting Principles Board largely adopted the all-inclusive income concept when it issued APB Opinion No. 9, Reporting the Results of Operations, and later reaffirmed the concept when it issued APB Opinion No. 20 and APB Opinion No. 30. Application of these pronouncements results in the presentation of discontinued operations, extraordinary items, and the cumulative effect of a change in accounting principle on the face of income statement (net of their related tax effects) immediately below income from continuing operations on the face of the income statement.

Although the FASB generally follows the all-inclusive concept of income adopted by the APB, it has occasionally made specific exceptions by requiring that certain changes in assets and liabilities bypass the income statement in the period they are recognized. Instead of being reported in income, these unrealized items are to be reported as elements of stockholder's equity on the balance sheet. Statements that contain these exceptions include SFAS No. 52 Foreign Currency Translation, SFAS No. 87 Employers' Accounting for Pensions, and SFAS No. 115 Accounting for Certain Investments in Debt and Equity Securities.

FASB issued SFAS No. 130 in response to users' concern over these items bypassing the income statement and appearing only in the statement of changes in stockholders' equity. The information provided by comprehensive income was expected to assist investors, creditors and other financial statement users in evaluating an enterprise's economic activities, and its timing and magnitude of future cash flows. However, the disclosure of comprehensive income created an additional performance measure that many feared would confuse readers and would prove more volatile than net income (Hirst, 2006). Another major criticism of SFAS No. 130 is that the resulting comprehensive income figure is incomplete. Given the FASB's partial approach to fair value accounting, these OCI items capture some fair value changes for assets but disregard liability fair value changes (Hirst, 2006). While SFAS No. 130 mandates the reporting of these OCI items, it does not unify the presentation of them due to the allowance of three reporting alternatives.

REPORTING ALTERNATIVES

The first alternative uses a combined statement of net income and comprehensive income. Companies that elect to use this method report comprehensive income items at the bottom of the traditional income statement after net income. The advantage of this approach is that both measures of the entity's performance, net income and comprehensive income are disclosed in a single statement. Thus, users of the financial statement are less likely to miss OCI items in their decision making process. The primary disadvantage is that net income can be looked at as a subtotal in the income statement and comprehensive income can be thought of as the new bottom line. This will reduce the prominence of net income as the principle measure of a company's performance and may cause confusion among some financial statement users about true earnings (Campbell et al., 1999). However, the confusion should occur for a short period of time during the implementation of the standard for unsophisticated users because if the FASB chooses to enforce this format, even unsophisticated users will grow accustomed to the format.

The second method for reporting comprehensive income uses a separate financial statement. The statement begins with net income and concludes with comprehensive income. One advantage of this approach is that the income statement is kept free of potentially distracting disclosures about comprehensive income. Companies that view net income as the more meaningful performance may elect this approach because it does not change the income statement. Also the separate comprehensive income statement that is reported helps sophisticated professional investors who can utilize the additional information. The primary disadvantage of this approach is that it creates another statement, adding to the four traditional financial statements (Campbell et al., 1999). However, if companies must report OCI items in a certain format to comply with FASB's pronouncement, the cost of issuing one more financial statement will be minimal and users will be accustomed to the financial statement after some period of time.

The third approach reports comprehensive income in the statement of stockholders' equity. For most companies, this approach will be the closest to prior practice. The statement of stockholders' equity is the place where all of the components of comprehensive income have been previously shown. To comply with SFAS No. 130 using the third approach, companies only need to show how these components are added together to produce comprehensive income and add disclosures about tax effects. The primary advantage of using this approach is that companies can soften the appearance of comprehensive income as a performance measure. A potential disadvantage exists for companies that have previously relegated the statement of stockholders' equity to the footnotes. Because the FASB requires that the statement disclosing comprehensive income be given the same prominence as other financial statements, companies that choose to disclose comprehensive income in the statement of changes in stockholder's equity will no longer be able to put the statement in the footnotes (Campbell et al., 1999).

FASB does not mandate any one of the three possible financial statement formats for reporting comprehensive income. However, the Board encourages reporting entities to show the components of OCI and total CI in either a combined statement of net income and comprehensive income or in a separate statement. Regardless of the format used, comprehensive income per share is not shown and earnings per share will continue to be based on net income. Cumulative total OCI for the period should be presented on the balance sheet as a component of stockholders' equity, separate from additional paid in capital and retained earnings.

PRIOR RESEARCH

Several empirical and survey-based articles have examined the importance of comprehensive income and the preference of reporting. King et al. (1999) surveyed chief financial officers (CFOs) of publicly traded companies prior to the effective date of SFAS No. 130 to determine which of the three reporting formats the CFOs intended to use and whether the CFOs considered reporting comprehensive income useful to financial statement users. Approximately 67% of the surveyed CFOs stated that they preferred the option of reporting comprehensive income in a statement of changes in stockholders' equity while 33% preferred one of the two performance-based financial statement formats. In addition, a majority of the CFOs indicated that reporting comprehensive income was either not useful (35.9%) or actually misleading (38.5%) to users. They found a strong correlation between the respondents questioning the usefulness of reporting comprehensive income and the preference for reporting OCI items in a statement of changes in stockholders' equity. In addition to examining CFOs beliefs and intentions they also surveyed the professional users of the financial statements to determine their preferences in reporting format. Contrary to CFOs, 82% of the users preferred that comprehensive income be reported in one of the two performance-based financial statements. Only 18% preferred reporting in a statement of changes in stockholders' equity. Also, the format of reporting comprehensive income appeared to have an impact on whether these analysts would use comprehensive income in computing traditional performance measures such as return on equity. Reporting comprehensive income in a statement of changes in stockholders' equity lessened the likelihood that it would be used in computing performance ratios.

Hirst and Hopkins (1998) reached a similar conclusion in an experiment conducted with professional security analysts and portfolio managers. They examined one component of OCI, unrealized gains and losses on available for sale securities, and found that displaying this information in one of the two performance-based financial statements (as originally proposed in the Board's exposure draft) was effective in revealing to the professional investors a company's active earnings management through its marketable securities portfolio. Displaying the information in a statement of changes in stockholders' equity (as finally allowed in SFAS No. 130) was not effective in revealing this type of active earnings management to the users. Maines and McDaniel (2000) investigated the issue from the standpoint of non-professional investors. They conducted an experiment with individual investors, and their results showed that non-professional investors would use comprehensive income information in evaluating management performance only if it is presented in a separate statement of comprehensive income.

More recently, Hunton et al. (2006) conducted an experiment using financial executives and chief executive officers and found that subjects tended to buy or sell securities to manage earnings to achieve earnings forecasts. The use of a more transparent format (separate statement) for reporting comprehensive income significantly reduced this behavior. Subjects in the less transparent format (stockholders' equity statement disclosure) indicated these earnings management attempts would not be easily detectible by readers. Subjects in the more transparent format indicated these attempts at earnings management would be easily detectible by readers. Lee et al. (2006) sampled firms in the property-liability insurance industry and found that insurers with a tendency to manage earnings through security sales and insurers with reputations for poor disclosure quality are more likely to report comprehensive income in the statement of stockholders' equity.

Thus, all these studies examining the usefulness of comprehensive income in relation to its reporting format reached similar conclusions; placement of comprehensive income in a performance-based versus nonperformance-based financial statement signals the importance of comprehensive income information to users and impacts their use of this information. Reporting comprehensive income in a statement of changes in stockholders' equity conveys to users that this information is unrelated to corporate performance and therefore, is used little by investors. Moreover, disclosure in the statement of stockholders' equity can be an aid to firms who wish to manage earnings without detection.

SIGNIFICANCE OF OCI ITEMS

Campbell et al. (1999) examined the 1997 financial statements of 73 companies that adopted SFAS No. 130 early. They found that the average impact of OCI relative to net income was material and positive for those companies that chose the formats of the combined statement of net income and comprehensive income or the separate statement of comprehensive income as FASB recommended. Companies that chose the combined statement format had OCI that was, on average 57% of net income. Those that chose the separate statement format had average OCI that was 81% of net income. As a result, comprehensive income was substantially higher than net income in both of these groups. In contrast, the firms that chose the statement of stockholders' equity format had a material negative amount of OCI, averaging 17% of net income.

Jordan et al. (2002) studied a sample of 100 randomly selected financial services firms for 1998. The study also revealed the significant effects of OCI items compared to net income. Using a materiality threshold of 10%, 54 firms reported a material amount of OCI. Among them 11 firms reported OCI that was more than 100% (either positive or negative) of the net income. Even though the study was limited in scope due to the same type of firms being studied for a single year, it demonstrated that the significance of OCI in evaluating companies' operating performance potentially should not be ignored. If OCI are significant and different placement of reporting OCI items affects visibility and usefulness to financial statement users, FASB should consider eliminating the option of reporting OCI in the statement of changes in stockholders' equity.

SAMPLE FIRMS AND DATA COLLECTION

The Fortune 500 companies were chosen for analysis in the current study. These large firms are likely to have the type of transactions that would be captured in other comprehensive income (OCI) and not net income. In addition the Fortune 500 firms consist of companies in a wide range of industry classifications. Previous studies have been limited in the number of firms or the type of firms analyzed. Using the Fortune 500 as a sample overcomes these limitations of previous studies.

The Fortune 500 list has chronicled big business in the United States since it was first compiled in 1954 (Clifford, 2001). Revenue has remained Fortune's constant criterion for ranking the largest companies. The 2000 list was the initial year included in this study and was based on operating results for 1999. As the first Fortune 500 of the 21st century, the 2000 list included such notable firsts as: the first pure internet company to make the list--AOL; first woman CEO to make the list--Carly Florina of Hewlett-Packard; and first biotech company to make the list--Amgen (Watson et al., 2000).

The 2001 list saw Exxon Mobil overtake General Motors as the largest U.S. company for the first time since 1984. Higher oil prices helped energy giants Duke Energy and Reliant Energy nearly double their revenues, and paved the way for the rise of diversified energy companies like Enron and Dynegy (Clifford, 2001). Of the 59 new arrivals on the 2001 list, twelve were from the energy industry classification. Other industries with significant increases included hotels and casinos (ten), pipelines (nine), and rubber and plastics (eight). Industry classifications with significant decreases included specialty retailers (ten), food (ten), motor vehicles (nine).

The 2002 list included 44 new firms. However, unlike the 2001 list, the industry classification totals remained stable with no industry gaining or losing more than three companies. The turnover of 59 companies (11.8%) in 2001 and 44 companies (8.8%) in 2002 approximate the 10% to 20% annual rate predicted by Fortune when the list was introduced (McLean, 2000). We found no evidence that inclusion in the Fortune 500 list affected valuation of the firms. However, Fortune unveiled two new stock indexes during the time period of this study (McLean, 2000). The first is based on the Fortune 500 list and the second, Fortune e-50, is based on Fortune's list of the 50 companies that best reflect the internet revolution. The Fortune 500 Index is designed to measure the stock performance of the largest U.S. businesses. Much of the stability of the Fortune 500 list itself comes from the fact that companies are ranked by revenue and not by more volatile factors like market value or earnings (McLean, 2000).

For each year from 1999-2001 financial statements were reviewed from SEC filings and/or company websites. These years build upon studies conducted on early adopters in 1997 and studies conducted on initial reporting of comprehensive income in 1998. Data was collected on the industry classification, method utilized to report comprehensive income, net income, components of OCI, and comprehensive income for each firm.

REPORTING METHOD UTILIZED

Results in Table 1 show that disclosure in the statement of stockholders' equity is the clearly favored choice of reporting method by the Fortune 500. The data for the three years 1999 to 2001 reveal that 69%, 68.4% and 74.2%, respectively, chose this method. These figures are slightly higher than those reported in earlier studies and indicate a small increase over the three years. The next most popular method is the separate statement of comprehensive income. The data show that 14.6%, 12.4%, and 16% of the Fortune 500 used the separate statement in 1999, 2000 and 2001. This shows a fairly steady number of firms choosing this method. The combined statement was chosen the least often as the reporting method each and the number of firms using this method declined steadily over this time period. Perhaps the most interesting finding was the surprising number of firms that did not report comprehensive income. The firms not reporting comprehensive income jumped from 13% in 1999 to 17% in 2000, and dropped dramatically to 7.8% cent in 2001. The dramatic drop in 2001 may be partially attributed to the 44 firms that failed to make the list again and the 44 new firms added. Fifteen of the 44 dropping off the list did not report comprehensive income information for 2000 while only three of the 44 new firms did not report comprehensive income information for 2001 operating results. A possible explanation for the remaining difference is materiality. OCI as a percentage of net income was greater than a positive or negative 3% for 352 firms in 2001 and for only 291 firms in 2000 (see Table 3). More firms may have chosen not to report detailed comprehensive income information in 2000 because OCI items did not materially affect their financial statements.

IMPACT OF OCI ITEMS

The most dramatic impact of OCI items for a company occurs when the two performance measures (net income and comprehensive income) have different signs. Table 2 shows the number of instances where this occurred each year. OCI turned a net loss into positive comprehensive income for no firms in 1999, 4 firms in 2000, and 2 firms in 2001. OCI turned a net income into a comprehensive loss for 19 firms in both 1999 and 2000, and 24 firms in 2001. This indicates that other comprehensive income is more likely to negatively affect performance than to enhance it.

Tables 3 and 4 also bear out this conclusion. Table 3 examines the relationship between OCI and net income. The total OCI for each firm was divided by the absolute value of the net income to determine the direction and percentage impact of OCI on net income. The table is arranged in gradients of materiality (positive and negative 2%, 3%, 5%, 10%, and 100%) with zero or not reported as the anchor. Note that the number of firms with zero comprehensive income or not reported in Table 3 is greater than the numbers for "not reported" in Table 1 because some firms reported zero comprehensive income while others did not disclose any comprehensive income information. The number of firms in the negative gradient of materiality in Table 3 is greater than the number of firms in the corresponding positive gradient of materiality in all cases for each of the three years except one. That case is occurs in 1999 (up to 1.9%--52 firms, compared to up to -1.9%--47 firms). In each year, the total number of firms negatively impacted by OCI is greater than the number of firms positively affected.

Table 4 tracks net income, other comprehensive income, and comprehensive income for each of the three years examined. It also shows the overall impact of OCI in relationship to net income. OCI was negative each year and trended downward sharply. The ratio of OCI to net income for the sample was -1.9% for 1999, -3.4% for 2000, and -30.9% for 2001. The modest decrease from 1999 to 2000 was due to the large increase in net income that partially offset the even more dramatic decrease in OCI. The sharp decrease in OCI to net income from 2000 to 2001 was caused by the large drop in net income coinciding with the large increase in negative OCI.

COMPONENTS OF OCI

Table 5 tracks six components of OCI for the three years. Foreign currency translation adjustments are the most significant component of OCI for each year. And for each year the impact of the foreign currency translation is negative. Unrealized gains/losses on marketable securities and minimum pension liability adjustments were more volatile, shifting from large positive amounts in 1999 to negative amounts in 2000, and then to even larger negative amounts in 2001. Reclassification adjustments remained a fairly consistent negative amount over the three years. This indicates that companies realized gains in each year that had previously been included in OCI. Income taxes and minority interest was a volatile category changing from a positive figure in 1999 to a negative amount in 2000, and to an even larger negative amount in 2001. The "other" items were also somewhat volatile, though relatively small in amount. These items changed from negative in 1999 to positive in 2000 and back to negative in 2001.

SUMMARY AND CONCLUSIONS

Since the requirement of reporting comprehensive income and its components by the FASB took effect in 1998, concern has arisen over the impact these items would have on the financial statements. Early trends in reporting comprehensive income and its components for the Fortune 500 reveal an overwhelming preference for disclosure in the statement of changes in stockholders' equity, despite the FASB's recommendation of utilizing a combined statement of income/comprehensive income or a separate statement of comprehensive income. This disclosure tends to downplay the importance of other comprehensive income items and focus readers' attention on the traditional net income figure rather than comprehensive income. Data from the Fortune 500 show that OCI items can indeed be volatile and significant, increasing in impact from a -1.9% of net income in 1999 to -30.9% of net income in 2001. The most significant component of OCI was the foreign currency translation adjustment, which was negative in each year examined. Unrealized gains/losses on marketable securities and minimum pension liability adjustments tended to be large and volatile.

It is not uncommon for companies to disregard the expressed preference of the FASB in reporting under its standards. For example the indirect method is utilized predominately over the direct method for reporting cash flows from operations despite the FASB's stated preference for the direct method. The intrinsic method of calculating stock option expense was also utilized predominately over the fair market value method before FASB finally required fair market value accounting for stock options rather than merely expressing a preference for it. Perhaps it is time for the FASB to reconsider the reporting flexibility afforded companies under SFAS No. 130. Requiring the OCI items to be disclosed in a combined statement of income and comprehensive income or in a separate statement of comprehensive income would allow these volatile and potentially significant items to be evaluated more directly by users of the financial statements.

REFERENCES

Campbell, L. and D. Crawford and D. Franz (1999). "How Companies Are Complying With the Comprehensive Income Disclosure Requirements," Ohio CPA Journal, Volume 58, Issue 1 (January-March), 13-20.

Clifford, Lee (2001). "Fortune 5 Hundred," Fortune, Volume 143, Issue 8, (April 16), 100-103.

Financial Accounting Standards Board (1997). Reporting Comprehensive Income. Statement of Financial Accounting Standards No. 130. Stamford, CT: FASB.

Hirst, D. Eric. (2006). "Discussion of 'Cherry Picking, Disclosure Quality, and Comprehensive Income Reporting Choices: The Case of Property-Liability Insurers'," Contemporary Accounting Research, Volume 23, Number 3 (Fall), 693-700.

Hirst, D. Eric and Patrick E. Hopkins (1998). "Comprehensive Income Reporting and Analysts' Valuation Judgments," Journal of Accounting Research, Volume 36, Supplement, 47-75.

Hunton, James E. and Robert Libby and Cheri L. Mazza (2006). "Financial Reporting Transparency and Earnings Management," The Accounting Review, Volume 81, Number 1, 135-157).

Jordan, Charles E. and Stanley J. Clark (2002). "Comprehensive Income: How Is It Being Reported And What Are Its Effects?" Journal of Applied Business Research, Volume 18, Issue 2 (Spring), 1-8.

King, T.E. and A.K. Ortegren and B.J. Reed (1999). "An Analysis of the Impact of Alternative Financial Statement Presentations of Comprehensive Income," Academy of Accounting and Financial Studies Journal, Volume 3, Number 1, 19-42.

Lee, Yen-Jung and Kathy R. Petroni and Min Shen (2006). "Cherry Picking, Disclosure Quality, and Comprehensive Income Reporting Choices: The Case of Property-Liability Insurers," Contemporary Accounting Research, Volume 23, Number 3 (Fall), 655-692.

Maines, Laureen A. and Linda S. McDaniel (2000). "Effects of Comprehensive Income Characteristics on Nonprofessional Investors' Judgments: The Role of Financial Statement Presentation Format," The Accounting Review, Volume 75, Issue 2 (April), 179-207.

McLean, Bethany (2000). "Introducing the FORTUNE Stock Indexes," Fortune, Volume 141, Issue 5, (March 6), 130-136.

Watson, Noshua, Garcia, and Feliciano (2000). "The Lists," Fortune, Volume 141, Issue 8, (April 18), 289-295.

Timothy L. McCoy, Lamar University

James H. Thompson, Washington State University

Margaret A. Hoskins, Henderson State University

Table 1: Method Used to Report Comprehensive Income and Its Components

 1999 2000

Reporting Method Number Percent Number Percent

Not reported 65 13% 85 17%

Combined Statement of 17 3.4% 11 2.2%
Net Income &
Comprehensive Income

Separate Statement of 73 14.6% 62 12.4%
Comprehensive Income

Included in Statement 345 69% 342 68.4%
of Stockholder's Equity

Total 500 100% 500 100%

 2001

Reporting Method Number Percent

Not reported 39 7.8%

Combined Statement of 10 2%
Net Income &
Comprehensive Income

Separate Statement of 80 16%
Comprehensive Income

Included in Statement 371 74.2%
of Stockholder's Equity

Total 500 100%

Table 2: Cases of Other Comprehensive Income Causing the Sign of
Net Income And Comprehensive Income to be Different

 1999 2000 2001

Firms with negative Net Income
 and positive CI 0 4 2
Firms with positive Net Income
 and negative CI 19 19 24

Table 3: Relationship of Other Comprehensive Income to Net Income

 1999 2000

OCI as % of NI Number Percent Number Percent

> 100% 9 1.8% 10 2.%
10% to 99.9% 47 9.4% 37 7.4%
5% to 9.9% 16 3.2% 19 3.8%
3% to 4.9% 15 3.% 13 2.6%
2% to 2.9% 14 2.8% 7 1.4%
Up to 1.9% 52 10.4% 25 5%
0 or Not Reported 70 14.% 100 20%
Up to -1.9% 47 9.4% 59 11.8%
-2% to -2.9% 18 3.6% 18 3.6%
-3% to -4.9% 25 5% 20 4%
-5% to -9.9% 45 9% 36 7.2%
-10% to -99.9% 120 24% 132 26.4%
>-100% 22 4.4% 24 4.8%
Total 500 100% 500 100%

 2001

OCI as % of NI Number Percent

> 100% 7 1.4%
10% to 99.9% 43 8.6%
5% to 9.9% 12 2.4%
3% to 4.9% 8 1.6%
2% to 2.9% 4 .8%
Up to 1.9% 26 5.2%
0 or Not Reported 49 9.8%
Up to -1.9% 54 10.8%
-2% to -2.9% 15 3%
-3% to -4.9% 37 7.4%
-5% to -9.9% 43 8.6%
-10% to -99.9% 157 31.4%
>-100% 45 9%
Total 500 100%

Table 4: Reported Net Income, Other Comprehensive Income, and
Comprehensive Income(in millions)

 Total NI Total OCI Total CI Total OCI/Total
 NI

1999 $445,516 ($8,414) $437,102 -1.9%
2000 $1,119,697 ($37,710) $1,081,987 -3.4%
2001 198,405 ($61,351) $137,054 -30.9%

Table 5: Components of Other Comprehensive Income(in millions)

Component of OCI 1999 2000 2001

Foreign Currency Translation ($20,714) ($25,704) ($19,471)
Unrealized G/L on
 Marketable Securities $10,484 ($245) ($8,817)
Minimum Pension Liability
 Adjustment $6,750 ($5,252) ($19,800)
Reclassification Adjustment ($5,249) ($5,190) ($6,492)
Income Taxes & Minority Interest 1,166 ($2,407) ($3,274)
Others ($851) 1,088 ($3,497)
Total OCI ($8,414) ($37,710) ($61,351)