Impairment Write-Offs: Truth or Manipulation.

The Asset Impairments Controversy

All entities, regardless of size, are subject to the risks and uncertainties of economic and technological change. Rapid technological advances, intense global competition, volatile interest and foreign exchange rates, and rapid changes in market demand can create obsolescence of plant, machines, and patents and can cause assets to lose some or all of their capacity to recover their costs. Poor management decisions on resource allocations can also impair the value of assets.

One view of write-offs of asset impairments is that management uses the accounting rules and manipulates earnings either by not recognizing impairment when it has occurred or by recognizing it only when it is advantageous to do so. An alternative point of view is that managers take write-offs not to manipulate earnings, but to reflect declines in the values of assets due to poor firm performance and poor management decisions. Francis, Hanna and Vincent [1] provide evidence that both factors (manipulation and impairment) can drive write-off decisions. However, when write-offs are analyzed by type; i.e., inventory, goodwill, property, plant, and equipment, and restructuring charges, they find that incentives play little or no role in determining inventory and fixed asset write-offs, but may play a substantial role in explaining other, more discretionary items, such as goodwill write-offs.

Both the frequency and the magnitude of write-offs increased during the late 1980s and in the 1990s, with financial press often expressing surprise when huge write-offs were reported. Many believed that the management and their accountants should have warned users of financial information about these write-offs and reported them over several periods. Therefore, it is not surprising that SEC Chairman Arthur Levitt criticized gimmicks such as big baths and the classification of merger costs as in process research and development, which can be written off immediately.

Moreover, write-offs do not occur randomly. The data used in this study shows that firms reporting write-offs tend to report additional write-offs every three years. The analysts refer to write-offs that evolve into a series of write-offs as "recurring/nonrecurring items," which alter the valuation implications of earnings and decrease the investors' confidence in their ability to value the entity. Since such beliefs erode confidence in financial reporting, the FASB responded by adopting Standard No. 121 (SFAS 121) in March 1995 which specified criteria for determining whether impairments of long-lived assets had occurred and how to measure and report them.

This article has two objectives. First, SFAS 121 is examined to determine if the recognition criteria and other requirements of the Standard can ensure that management accounts for asset impairments in a timely fashion. Second, some probable financial consequences of the Standard are examined through a simulation using publicly available data. The results indicate that the Standard makes it easier for management to report useful, reliable and timely information. If implemented effectively, the Standard should eliminate most of the periodic, large write-offs and improve the usefulness of certain popular financial ratios used in performance evaluation.

SFAS 121 Requirements

SFAS 121 requires the use of both undiscounted future cash flows and fair values to recognize and measure impairment. Paragraph five of SFAS 121 lists several impairment indicators, such as operating losses, significant adverse changes in the legal and economic climate and changes in the manner in which an asset is used. When one or more of these circumstances are present, an entity must estimate the net future cash flows (undiscounted and without interest charges) associated with the use of that asset and its eventual disposition. When estimating cash flows, assets are grouped at the lowest level at which cash flows are identifiable and independent from other groups. If these cash flows are less than the carrying amount of the asset(s), the entity should recognize an impairment loss.

The amount of impairment is dependent upon the fair value of the asset. SFAS 121, par. 7, states: "The fair value of an asset is the amount at which the asset could be bought or sold in a current transaction between willing parties, other than in a forced or liquidation sale." The statement goes on to explain that quoted market prices in active markets are superior to other valuation techniques, including the present value of future cash flows, (discounted in this step at a rate commensurate with the risk involved), option-pricing models, matrix pricing, option-adjusted spread models, and fundamental analysis. Once an asset to be held for use has been impaired, the carrying value may not be written back up, even if there is a change in circumstances.

Gains or losses on impairments of assets are reported in the income statement as ordinary income before income taxes, preferably under a separate caption. In addition, footnotes must include a description of impaired assets, the business segment(s) in which assets are located; the facts and circumstances leading to impairment; the expected disposal date, if applicable; and the carrying amount of these assets (pars. 13 and 19). The SFAS 121 is applicable to those assets which will be disposed of, by sale or abandonment, and which are not covered by current standards. The FASB is currently debating a new standard that will specifically address asset disposals and abandonment.

Can SFAS 121 improve Valuation and Reporting?

Can SFAS 121 force management to recognize, in a timely manner, asset impairments? Past controversies may be examined to assess the probable impact of SFAS 121 on their settlement. The recognition criteria allow management to use much judgement. Can impairment occur when the usual signs are not present? Most managers use both external criteria and internal calculations and determinations; e.g., net cash flow projections for an asset or a group of assets, inmost circumstances. Thus, management should implement policies and procedures to identify possible impairment indicators based on a company's operating environment.

At the second stage, the determination of net future cash flows calls for an estimate of the future cash inflows and outflows--undiscounted. Even for experts, such estimates may be subjective and difficult to accomplish. The same is true for exit values where no ready market exists. According to Braun, Rohan, and [Yospe.sup.2], another second stage issue is the aggregation level where net cash flows are determined. Suppose a group consists of two assets and that one asset becomes impaired while the other experiences an increase in value. Should they be allowed to cancel each other or should the first asset be disaggregated from the group with a recognized impairment?

An issue at the third stage is whether the impairment loss is limited to the unrecoverable cost of the asset. SFAS 121 uses a comparison of fair value to carrying cost to determine the impairment loss. One method of determining fair value is to discount the projected net cash flows the asset will generate. Discounting carries built-in profits for future periods by reducing future depreciation by an amount equal to interest charges that result in the impairment loss. The fourth stage, the disclosure of impairment, is not controversial and employs the traditional rules that exist for loss contingencies.

Finally, there are two general issues. First, there is a question of the implementation cost versus the benefit. While estimating future cash flows can be time-consuming and expensive, many companies already have capital budgeting policies that require review of assets to determine whether original projections are being met. Second, some companies report asset impairment only at year-end. However, quarterly reports are important to creditors and investors; if management is aware of potential asset impairment that might affect quarterly earnings, it should consider the appropriate time to do asset impairment tests and report impairment losses.

In summary, it appears that SFAS 121 has accomplished much in giving direction and establishing a framework for impairment accounting. However, two controversies that were instrumental in allowing management manipulation of write-offs and causing diversity in practice still remain. First, management may want to overestimate future cash flows to avoid write-offs. At other times, management may believe a significant write-off will benefit the company in the long-term. In these instances, they may underestimate cash flows. Second, unrealized gains of some assets could offset unrealized losses of others, resulting in no required write-off. Consequently, management may tend to group assets inappropriately at times to avoid asset impairment write-offs.

The Impact of SFAS 121

The effects on financial statements of SFAS 121 are three-fold. First, the income statement is affected by the amount of the impairment loss. Second, the balance sheet is affected in that net equity would decrease by the amount of the impairment through revaluation of the asset and the charge to retained earnings from the income statement. In future periods it is expected that stated asset values will more closely resemble actual market values, and earnings should normally increase as a result of decreases in depreciation expense. Third, impairment losses should decrease in size (fewer "big baths"), but reported more frequently, resulting in timely information.

A dissenting FASB member, Mr. Northcutt, argued that the impairment losses should be handled as changes in accounting estimates and recorded through revisions in the depreciation rates by adjusting the useful lives and salvage values of assets prospectively. He suggested that managements employ value-in-use measurements and frequently evaluate asset values, useful lives, salvage values, and depreciation or amortization amounts. We believe that the Standard will ultimately force management to follow the process of annual or quarterly evaluation of depreciation or amortization estimates because it will be more cost effective and routine to do so, substantially eliminating the need for periodic write-offs. The ever increasing pace of change in our economy and technology is gradually reducing the useful lives of all assets, lending credibility to this prediction.

To determine some probable financial consequences of SPAS 121, financial ratios commonly used to evaluate managerial competence and investment returns were analyzed by using the return on total assets (ROA) and return on common equity (ROE) measures of a number of companies. The 1995, 1996 and 1997 annual reports of 1,238 industrial and retail companies included in the Moody's Industrial Manuals were examined. Only 208 (16.8%) of these companies reported impairment losses during this period. Industrial and retail companies were targeted because they owned large amounts of fixed assets and intangibles, making it more probable that they would have impairment losses. The 1995-1997 time period was selected to ensure that study included impairment disclosures that occurred during the years immediately preceding and following the issuance of the Standard. Finally, a three-year time span was selected because the data analysis indicated that most companies reporting impairments reported them often and, on the averag e, in three-year intervals.

Using the management discussion and analysis, financial statements and footnotes of these 208 annual reports, 45 companies (3.4%) were chosen that had material asset impairment losses in at least one of these three years. While most of the 208 losses disclosed in the footnotes involved restructuring and personnel costs, which made them unsuitable for our analyses, 45 instances selected involved impairments of fixed assets and intangibles. SPAS 121 were mentioned in 19 of these 45 cases, ensuring that the post-Standard period was represented in the study. The companies identified were in paper, petroleum, chemicals, rubber, consumer products, electronics, and retail sectors.

The review of the impairment footnotes resulted in two important observations. First, contrary to popular anecdotal evidence concerning footnotes that one encounters in the print media and at professional meetings, the discussions of impairment charges were informative and clearly written. For example, Hasbro Corporation reported a $140 million non-recurring loss in its 1997 income statement. A careful examination of the footnotes revealed that only $20 million of this amount was an impairment loss. The remaining $120 million included losses from disposal of a segment of business, personnel relocation costs, pension costs, and other restructuring charges. Similarly, footnotes that clearly described the related transactions allowed for the identification of the majority of the initial 208 impairments selected as restructuring and employee termination costs--reducing the number of observations involving FASB 121 type asset impairments to 45. Second, the majority of the impairment observations generally follow ed a three year pattern.

Using the information presented in the 45 annual reports, ROA and ROE values were computed for the year in which the impairment loss occurred and for the two preceding years. Next, a simulation was undertaken. The two-thirds of the observed impairment loss (net of effective tax rate disclosed in footnotes) was added back to or subtracted from the net income (or loss) disclosed in the annual report, while one-third each was subtracted from (or added back to) the net income (or loss) disclosed in the annual reports of the preceding two years. The choice of the three-year allocation period was consistent with the observed data. Using these hypothetical net income amounts, ROA and ROE ratios were recomputed.

Assuming that the procedures described above effectively simulated the impact of SPAS 121 (more frequent and smaller write-offs, fewer "big bath?) on the ROA and ROE ratios, tests were conducted to determine if the differences between the actual and simulated amounts for each year were statistically significant. Given that each company had three paired (actual/hypothetical) observations for each ratio over the 1993-1997 period (any three out of the five years examined -- 1993-95, 1994-96, or 1995-97), the appropriate statistical procedure to test the level of the significance of the differences was a one-tailed t-test. This test would show if the increases in the ROA and ROE measures in the year of the add-back and the decreases in the same measures during the preceding two years were significant with a 95 percent confidence level. Since the test hypothesis is usually stated in a form indicating no differences (the null form) between the means of the paired observations, a probability of less than .05 shows that the impact of SFAS 121 is significant on the companies included in that group.

Table 1 shows that all t-statistics are close to zero, indicating with 99% or higher confidence that the simulated impact of SFAS 121 is significant on these ratios. Thus, if companies included in this study implemented SFAS 121 by evaluating their long-term assets quarterly or annually and by adjusting their depreciation estimates prospectively over a three-year period, there would be significant impact on the two principal management competence ratios. This result may mean that, in the long run, SFAS 121 should lead to the disclosure of more timely and credible information, with the assumption that impairment losses would be accrued each year through depreciation charges.

Concluding Comments

One of the primary reasons that SFAS 121 was issued is to increase the consistency and comparability of financial statement information among entities. The SFAS 121 requirements for asset valuation make it clear that management must evaluate asset values quarterly or annually, as appropriate. Another benefit should be an increase in consistency and comparability due to the uniformity that SFAS 121 brings to the calculation of the impairment of long-lived assets. The fact that the Statement makes greater use of forward-looking information, projected cash flows and fair values helps to make the information contained in financial statements more useful.

Some inconsistencies may remain due to the amount of subjective estimates and judgement involved in applying the Standard, but they should be occurring in a greatly reduced rate when compared to the situation before its issuance. Despite the amount of "room" left in the Standard that may allow management to manipulate the figures, we believe that the application of SFAS 121 will provide users with better insight into management policies and actions. If implemented effectively, the Statement may have financial consequences different from those currently reported, resulting in a more gradual but frequent accrual of impairment losses through depreciation or other charges.

References

(1.) Francis, J., J.D. Hanna and L. Vincent. 1996. "Causes and Effects of Discretionary Asset Write-offs." Journal of Accounting Research; 34 (Supplement): 117-134.

(2.) Braun, S., P. Rohan and J. F. Yospe. 1991. "Asset Write-offs: A Matter of Grouping? Journal of Accountancy; 171 (4, April): 63-68.

Paul D. Baker is Professor of Accounting at the State University of West Georgia at Carrollton, Georgia. He has a Ph.D. in Business Administration (major field--Accounting), a Master of Business Administration, and a Bachelor of Business Administration (major--Accounting) from Georgia State University. His teaching experience includes the University of South Carolina and Georgia State University. He is the author of numerous articles and published papers.

Joseph C. Rue received his Ph.D. from Penn State and is currently serving as Professor of Accounting at Florida Gulf Coast University. He has worked for Ernst and Young and Stockton Bates and Co. as a public accountant. Dr. Rue's research interests are primarily in financial accounting. Recent publications have focused on pension accounting and deferred taxes, including articles in the National Public Accountant.

Ara G. Volkan received his Ph.D. in accounting from the University of Alabama. He is a Florida and Georgia CPA. He is currently the Chair of the Accounting and Finance Department at the State University of West Georgia with the rank of Full Professor. Dr. Volkan has published numerous articles, including articles in the National Public Accountant, and conducted seminars for various academic and professional organizations.