Download Big Bath Earnings Management: Evidence under SFAS No. 142 and more Lecture notes Accounting in PDF only on Docsity!
Big Bath Earnings Management:
The Case Of Goodwill Impairment
Under SFAS No. 142
Charles E. Jordan (Email: Jordan@cba.usm.edu), University of Southern Mississippi Stanley J. Clark (Email: Clark@cba.usm.edu), University of Southern Mississippi
Abstract
The big bath theory of earnings management suggests that firms experiencing low earnings in a given year may take discretionary write downs to reduce even further the current period’s earn- ings. The notion is that the company and its management will not be punished proportionately more for the big hit it takes to its already depressed earnings. This “clearing of the decks” makes it easier to generate higher profits in later years. SFAS No. 142, with its new requirement to test goodwill annually for impairment, provided a unique opportunity to test this big bath theory. Ex- amining Fortune 100 companies, this study presents compelling evidence that the big bath theory is more than just a theory but is instead a practiced method of managing earnings.
1. Introduction
ecent corporate scandals and accounting improprieties at major companies, such as Enron, World- Com, Waste Management, Sunbeam and many others, have shaken investor confidence in the financial reporting process. This current demise in investor confidence was foreseen several years ago by former Securities and Exchange Commission (SEC) chairman Arthur Levitt. In a September 1998 speech, then chairman Levitt stated, “In the zeal to satisfy consensus earnings estimates and project a smooth earnings path, wishful thinking may be winning the day over faithful representation. As a result, I fear that we are witnessing an erosion in the quality of earnings and therefore the quality of financial reporting (Springsteel, 1998, p. 21).”
R
In his almost prophetic speech, Mr. Levitt was referring to the practice of earnings management, which embodies deliberate steps taken within generally accepted accounting principles (GAAP) to bring about a desired outcome. Earnings management can be accomplished because GAAP-based financial statements require the use of many estimates and judgments (e.g., estimating the useful lives of plant assets, determining whether assets have been impaired, or deciding upon amounts accrued for loss contingencies just to name a few). One subset of earnings management involves “big bath” charges, which represent significant non-recurring losses or expenses taken in the current period to clear the decks for improved future earnings performance (Sikora, 1999).
In June 2001, the Financial Accounting Standards Board (FASB) issued Statement of Financial Accounting Standard (SFAS) No. 142, Accounting for Goodwill and Other Intangible Assets , and created the potential for big bath earnings management in relation to goodwill impairment. In particular, SFAS No. 142, which became effective in 2002, eliminates the periodic amortization of goodwill but instead requires that goodwill be evaluated each year for impairment. If impairment exists, an immediate charge to earnings must be recorded for the amount of the impairment. Testing goodwill for impairment under SFAS No. 142 involves significant use of estimates and, thus, opens the door for earnings management. For a sample of companies, the current study examines whether the recording of goodwill impairment in the year of adopting SFAS No. 142 appears to be related to the big bath theory of earnings management.
2. Literature Review
Dye (1986) notes that management has two primary reasons to manage or manipulate earnings. One is an external demand to meet earnings forecasts and increase share price; the other represents an internal demand relating to optimal contracting. In this latter case, earnings allow managers to communicate with their principals (e.g., board of directors) concerning the level of their performance. Regarding the external demand to meet earnings forecasts, Chenheiter and Melumad (2002) note that, ceteris paribus , investors infer a higher level of permanent cash flows from a higher level of reported earnings. Since increasing cash flows translate into higher share prices and earnings are perceived to be a surrogate of cash flows, higher earnings increase the value of the firm. Numerous studies have found positive evidence of earnings management being used either to meet earnings forecasts or to promote optimal contracting (e.g., see Healy, 1985; Moses, 1987; Trueman and Titman, 1988; Fudenberg and Tirole, 1995; Jordan et al., 1997/1998; Lu, 2000; Schrand and Wong, 2000).
However, managing earnings through big bath charges follows a different, yet simple, line of reasoning be- cause earnings are made to look worse, at least in the current period. Henry and Schmitt (2001) note that a company will take a large non-recurring loss one year, typically when its profits are already depressed, so that future earnings are not burdened. The result is either increased future earnings or reduced variability of future earnings. The notion is that, when things are already bad (i.e., depressed earnings), making them worse by clearing out the rubbish does little harm to the company’s or management’s reputations. The market punishes a firm relatively the same whether it misses its earnings mark by a little or by a lot.
Although the big bath theory has been espoused in the accounting literature for years, little empirical test- ing of its presence exists. A few studies examined big bath charges on the periphery or as an aside to their main topic of earnings management in general. For example, Cameron and Stephens (1991), in examining the impact of non-recurring items on the predictive ability or variability of earnings, found that these items are not used to smooth earnings but instead appear to be used more consistently with the big bath theory. Bauman et al. (2001) examined earnings management in relation to the discretionary adjustments associated with the valuation allowance for deferred tax assets and found virtually no evidence in support of earnings management. They did find that firms with negative earnings tended to book significant negative adjustments to the valuation allowances, which is consistent with the big bath theory. However, they warn that the big bath theory may not be driving this result as it could be nothing more than companies applying the rules of SFAS No. 109 the way they were intended. More specifically, a net loss in the current period is one of the signs the FASB states may indicate that sufficient future taxable income to realize the benefits of the deferred tax asset may not materialize. Thus, negative earnings in the current period represent a legitimate reason to increase the valuation allowance. Yoon and Miller (2002), in a study of Korean industrial firms, provide significant evidence that the level of operating performance affects the degree of earnings management. They also found evidence supporting the big bath theory in that, when operating perform- ance was extremely poor, firms often took income-decreasing strategies rather than income-increasing ones.
Four studies examined big baths or large write downs as a primary point of investigation. Strong and Meyer (1987) performed a capital market study in relation to announcements of major asset write downs. They did not address write downs in relation to earnings management but instead concluded that the most important determi- nant of a write down decision is a change in senior management, especially if the new chief executive comes from outside the company. Elliott and Shaw (1988) arbitrarily defined a big bath as a write down exceeding 1 percent of the book value of a firm’s assets. Their mainly descriptive results showed that companies taking these discretionary big baths tended to be larger than other firms in their respective industries and more highly leveraged as well; they also seemed to be underperformers in terms of earnings.
For a sample of Australian firms, Walsh et al. (1991) examined large losses and large gains reported as ex- traordinary items. Their results showed a strong correlation between the discretionary loss or gain reported as an extraordinary item and the level of current year earnings. Companies with unusually low current year earnings in relation to prior years were more prone to take large discretionary losses as extraordinary items, while firms with unusually high current year earnings tended to recognize large discretionary gains as extraordinary items. Chen- heiter and Melumad (2002) provide additional evidence concerning earnings management and big baths. In their
among firms. However, it may be inappropriate for service firms, which have lower levels of total assets than either manufacturing or retail firms. Thus, ROS was also examined as it eliminates the bias that may be present in ROA computations. Medians were used as summary measures for the groups rather than means because means can be unduly influenced by a few extreme observations, especially for small sample sizes like those examined in the current study. Medians are much less affected by these extreme values.
4. Results
Of the Fortune 100 companies examined, 20 firms either did not report an amount for goodwill in 2001 or 2002 or had missing data. Of the 80 remaining companies that reported goodwill in their financial statements, 29 (36.3 percent) recorded a goodwill impairment loss in 2002 under SFAS No. 142 while 51 (63.7 percent) did not. For the 29 firms reporting an impairment loss, Table 1 provides information on the relative size or amount of the 2002 loss. Notice that the amount of the impairment loss appears significant. For example, the median loss to the amount of 2001 goodwill (i.e., goodwill before impairment) was 20.02 percent. The 75th^ percentile for this ratio is 72.45 percent, which indicates that one fourth of the firms wrote off the vast majority of their existing goodwill.
Table 1: Significance Of The 2002 Impairment Loss
Ratio 25 th^ percentile 50 th^ percentile (median)
75 th^ percentile
Impairment loss to 2001 goodwill 6.64% 20.02% 72.45% Impairment loss to 2002 operating income 4.54% 13.79% 95.00% Impairment loss to 2002 total assets 0.15% 1.01% 4.60% Impairment loss to 2002 sales 0.27% 1.51% 5.61%
In a survey of most of the empirical literature on materiality at the time, Holstrum and Messier (1982) con- cluded that an item’s effect on income was the most important factor in determining the materiality of that item. They concluded that a general consensus existed among most parties (i.e., auditors, preparers, and users) that items producing income effects greater than 10 percent are considered material. Table 1 reveals that the median impair- ment loss to 2002 pre-tax operating income (i.e., income before the impairment loss or other special components such as extraordinary items, changes in accounting principle, or discontinued operations) was 13.79 percent. According to Holstrum and Messier (1982), this suggests that the amount of the impairment loss for the group as a whole was material.
Table 2 provides the median ROA, ROS, and goodwill to total assets for 2001 and 2002 for both groups of firms (i.e., those recording an impairment loss in 2002 and those not recording an impairment loss in 2002). Notice that goodwill represented a sizable portion of total assets for both groups of firms in 2001 (i.e., before impairment) with median amounts of 4.81 percent and 6.48 percent for the impairment and non-impairment groups, respectively. The medians for the goodwill to assets ratio did not differ between these two groups in 2001 at a statistically significant level (i.e., α = .10). However, in 2002, subsequent to implementation of SFAS No. 142, not surprisingly the median goodwill to assets ratio for the impairment group declined while the same ratio for the non-impairment group increased; the medians differed significantly between the two groups in 2002.
The most important result in Table 2 lies in a comparison of the earnings levels for the two groups of firms. In 2001, prior to implementation of SFAS No. 142, the median ROAs for both groups were quite similar as were the median ROSs. Statistical tests of the differences between the medians for 2001 revealed no significant difference. However, in the year when the impairment write downs were recorded by the 29 firms (i.e., 2002), the impairment group reported median ROAs and median ROSs that were significantly lower than the respective medians for the 51 firms in the non-impairment group. Note that these profitability measures are pre-tax and based on operating income before the effect of any impairment loss.
Table 2: Profitability Of The Impairment And Non-Impairment Groups Of Firms
2001 2002 Median: Impairment Group
Non-Impairment Group α^ level^
Impairment Group
Non-Impairment Group α^ level Goodwill to total assets 4.81%^ 6.48%^ .2786^ 3.42%^ 9.11%^. Return on assets 3.56%^ 2.93%^ .2291^ 2.16%^ 3.80%^. Return on sales 6.40%^ 6.35%^ .5747^ 4.32%^ 7.09%^. Note: α level is the significance level for a test of differences between the impairment group median and the non-impairment group median within the same year.
This represents an important finding and strongly suggests the presence of big bath earnings management in 2002. More specifically, big bath theory holds that firms with depressed earnings are more likely to engage in discretionary write downs. The data in Table 2 reveal that the impairment group experienced significant reductions in earnings between 2001 and 2002 (i.e., the median ROA and ROS for this group declined precipitously between the years). The non-impairment group, however, enjoyed improved earnings levels between the years. As discussed in the previous paragraph, in the year of the impairment loss (i.e., 2002), the impairment group experienced earnings levels significantly lower than those of the non-impairment group. Thus, all the evidence indicates that the impair- ment group suffered from depressed earnings in 2002, and this could be a primary reason the managers decided to take the write downs in 2002.
Henry and Schmitt (2001) note that companies with negative earnings may be more prone to take big hits than companies with positive earnings. This is simply a special case of stating that firms with extremely poor earnings are more likely to take big baths. However, it also provides another test for the presence of big bath earnings management in the two groups of firms studied. In particular, if big bath earnings management exists, one would expect the impairment group to experience a higher incidence of negative earnings firms in 2002 when compared to the non-impairment group. Table 3 presents the number of firms with positive earnings and negative earnings for both groups of firms in 2001 and 2002.
Table 3: Negative Vs. Positive Earnings Firms For The Impairment And Non-Impairment Groups
2001 2002 Impairment Group Non-Impairment Group
Impairment Group Non-Impairment Group Number of firms with negative earnings 6 (20.7%)^ 6 (11.8%)^ 9 (31.0%)^ 4 (7.8%) Number of firms with positive earnings 23 (79.3%)^ 45 (88.2%^ 20 (69.0%)^ 47 (92.2%) Total 29 (100%) 51 (100%) 29 (100%) 51 (100%) Z value 1.075 2. α level .2825. Note: Z values and α levels are for two-tailed proportions tests to determine if the proportion of negative earnings firms differed between the impairment group and non-impairment group. One test was performed for 2001, and another test was performed for 2002.
Table 3 shows that the proportion of negative earnings firms for 2001 did not differ between the impair- ment and non-impairment groups at a statistically significant level. However, for 2002 when the impairment write downs were actually recorded, the impairment group had a significantly (α = .0069) higher rate of negative earnings firms than did the non-impairment group. Again, note that the earnings examined here are before any impairment losses. These results provide additional evidence suggesting that big bath earnings management occurred in 2002.
The current project examined big baths in 2002 partly because it was the year of adoption for SFAS No. 142 and also because it was the most recent year for which data were available. As time passes, however, later years can be examined to determine if companies with depressed earnings take big baths even though some of the incen- tive for doing so in relation to goodwill impairment no longer applies. More specifically, impairment losses sub- sequent to 2002 must flow through operating income and, thus, may carry a steeper penalty for firms than existed in the initial year of adopting SFAS No. 142. This may discourage big baths via goodwill impairment after 2002.
References
- Bauman, C., M. Bauman, and R. Halsey, “Do Firms Use the Deferred Tax Asset Valuation Allowance to Manage Earnings?,” Journal of American Taxation Association , Vol. 23, Supplement, pp. 27-48, 2001.
- Cameron, A. and L. Stephens, “The Treatment of Non-Recurring Items in the Income Statement and Their Consistency with FASB Concept Statements,” Abacus , Vol. 27, No. 2, pp. 81-96, 1991.
- Chenheiter, M. and N. Melumad, “Can Big Bath and Earnings Smoothing Co-exist as Equilibrium Financial Reporting Strategies?,” Journal of Accounting Research , Vol. 40, No. 3, pp. 761-796, 2002.
- Dye, R., “Proprietary and Non-proprietary Disclosures,” Journal of Business , Vol. 59, No. 2, pp. 331-366,
- Elliott, J. and W. Shaw, “Write Offs as Accounting Procedures to Manage Perceptions,” Journal of Accounting Research , Vol. 26, Supplement, pp. 91-119, 1988.
- Fudenberg, D. and J. Tirole, “A Theory of Income and Dividend Smoothing Based on Incumbency Rents,” Journal of Political Economy , Vol. 103, No. 1, pp. 75-93, 1995.
- Healy, P., “The Effects of Bonus Schemes on Accounting Decisions,” Journal of Accounting and Econom- ics , Vol. 7, No. 1-3, pp. 85-107, 1985.
- Henry, D. and C. Schmitt, “The Numbers Game,” Business Week , No. 3732, pp. 100-107, May 14, 2001.
- Holstrum, G. and W. Messier, “A Review and Integration of Empirical Research on Materiality,” Auditing: A Journal of Practice and Theory , Vol. 2, No. 1, pp. 45-63, 1982.
- Jordan, C., S. Clark, and W. Smith, “Earnings Management Under SFAS No. 115: Evidence from the Insurance Industry,” Journal of Applied Business Research , Vol. 14, No. 1, pp. 49-56, 1997/1998.
- Lu, J., “The Valuation Allowance for Deferred Tax Assets and Earnings Management,” Working paper, University of Southern California, 2000.
- Massoud, M. and C. Raiborn, “Accounting for Goodwill: Are We Better Off?,” Review of Business , Vol. 24, No. 2, pp. 26-32, 2003.
- Moses, O., “Income Smoothing and Incentives: Empirical Tests Using Accounting Changes,” Accounting Review , Vol. 62, No. 2, pp. 358-377, 1987.
- Schrand, W. and M. Wong, “Earnings Management and Its Pricing Implications: Evidence from Banks’ Adjustments to the Valuation Allowance for Deferred Tax Assets and SFAS 109,” Working paper, Univer- sity of Pennsylvania and University of California, Berkeley, 2000.
- Sikora, M., “Timing A Big Bath to an Acquisition,” Mergers & Acquisitions: The Dealmaker’s Journal , Vol. 33, No. 6, pp. 8-9, 1999.
- Springsteel, I., “The SEC Decries Special Charges,” CFO , Vol. 14, No. 11, p. 21, 1998.
- Strong, J. and J. Meyer, “Asset Write Downs: Managerial Incentives and Security Returns,” Journal of Finance , Vol. 42, No. 3, pp. 643-664, 1987.
- Trueman, B. and S. Titman, “An Explanation of Accounting Income Smoothing,” Journal of Accounting Research , Vol. 26, Supplement, pp. 127-139, 1988.
- Walsh, P., R. Craig, and F. Clarke, “Big Bath Accounting Using Extraordinary Items Adjustments: Australian Empirical Evidence,” Journal of Business Finance & Accounting , Vol. 18, No. 2, pp. 173-189,
- Yoon, S., and G. Miller, “Cash from Operations and Earnings Management in Korea,” The International Journal of Accounting , Vol. 37, No. 4, pp. 395-412, 2002.
Notes
