Voluntary Asset Write-Downs of SFAS 121
Early Adopters vis-à-vis Late Adopters
By
Brian Gaber
School of Administrative Studies
AtkinsonCollege
York University, Canada
416-736-5210 and 416-736-5963 (Fax)
Sungsoo Kim
School of Business
RutgersUniversity
Camden, NJ08102
856-225-6584 and 856-225-6231 (Fax)
Sung S. Kwon
School of Administrative Studies
AtkinsonCollege
York University, Canada
416-736-5210 and 416-736-5963 (Fax)
Current Version: March 15, 2006
Gaber and Kwon acknowledge the financial support from the Research Grant programof the School of Administrative Studies at YorkUniversity. Kim acknowledges the financial support from Research Grant Program of the School of Business at RutgersUniversity.
Voluntary Asset Write-Downs of SFAS 121
Early Adopters vis-à-vis Late Adopters
Abstract
Previous research shows that stock market positively values the firms whose motivation is to gain production efficiency through streamlining their assets as part of restructuring campaign. In this study, we examine 47 firms that voluntarily disclosed asset write-down information in either 10-K or ARS one year prior to the mandatory adoption of SFAS. No. 121.
We document the following empirical evidence: firstly, EARLY firms (those who adopted SFAS121 in 1995, one year prior to the mandatory adoption), pursuing production efficiency through restructuring efforts, experience more positive market reaction than LATE (those who adopted SFAS 121 in 1996) firms at the disclosure of assets write-down decisions, consistent with efficiency enhancement arguments. This evidence is robust to the selection of an alternative control group, matched in term of asset size and industry classification codes. Secondly, EARLY firms incur more capital expenditures, acquire less intangible, and exhibit lower asset turnover ratio and shorter asset age than LATE firms. Finally, EARLY firms show higher profitability, and have higher effective tax-bracket than those of LATE firms when they made voluntary decision of asset write-down. Previous studies that have examined the issue of asset write-down cross-sectionally aggregate the sample firms, and thus fail to recognize the different motivation of different firms. In this paper, we enhance our analysis by employing a voluntary disclosure sample, a control sample which made their asset write-down disclosures in the first fiscal year of mandatory adoption, and another control sample (MATCHED CONTROL), which did not make any asset write-down disclosures in either the voluntary year (1995) or the mandatory year (1996). We hope that this paper sheds light on the determinants that motivate firms to voluntarily disclose the asset write-down information in their 10-K reports or annual reports to shareholders one year prior to the mandatory adoption of SFAS. No. 121.
Keywords: Asset Write-Down, Early Adopters, Late Adopters, Efficiency Enhancement
Data Availability: The data used in this study are publicly available from the sources identified in the text.
Voluntary Asset Write-Downs of SFAS 121
Early Adopters vis-à-vis Late Adopters
I. Introduction
Given the latitude with which manager exercise in determining the extent and timing of the asset write-down, it is reasonable to expect that researchers observe empirical regularities in managers’ asset write-downs. There is mixed anecdotal evidence on the effect of the asset write-down. A group of firms reports positive price reaction to the asset write-down (IBM, Phillip Morris, Xerox, Upjohn), while other firms experience a decline in stock price immediately after the asset write-down (GTE, Kmart, Borden and Food Lion) (Kieso & Weygandt, 9th edition, p. 145). The body of prior literature failed to recognize the different motivation by different managers by cross-sectionally aggregating the sample firms. The market is expected to react positively to firms with lower asset turnover ratio streamlining their assets.[1] Asset write-down in this case will be viewed as part of efficiency enhancement consideration.
If well-performing firms involve in a discretionary asset write-down, the market will notice the managers’ opportunistic behavior and reactnegativelyto these non-value maximizing activities. (Zucca and Campbell, 1992, pp 30-41). SFAS No. 121, “Accounting for the Impairment of Long-Lived Assets and for Long-Lived Assets to Be Disposed of,” (hereinafter SFAS 121) poses an unique dilemma to managers in dealing with their asset impairment problems. Although they retain their discretion over the extent to which the assets are impaired under SFAS 121, they now have only limited discretion as to the timing of the asset write-down. Since the mandatory adoption of SFAS 121 began with the fiscal year ending 1996, they have a timing choice of either 1995 or 1996 to write down assets and recognize their impairments. Therefore, the time is running out for them to differentiate themselves from the rest of the firms if they desire to demonstrate that they engage in an efficiency enhancement activity.
In this paper, we first investigate the market effects of asset write-down by EARLY(those who adopt in 1995) versus LATE (1996 adopters) firms of SFAS 121 on asset impairments. Second, we attempt to explain why a group of firms elects early adoption of SFAS 121. There is a stream of literature that is somewhat related to our work. These articles study the association between prices and write-downs of capital asset book values (i.e., recognizing the partial impairment of assets that have not been retired). The results are mixed. Elliot and Shaw (1988) demonstrate an unfavorable market reaction to write-downs; Strong and Meyers (1987) show positive price reactions; and Francis et al. (1996) find mixed reaction. Specifically, Francis et al. (1996) document that investors respond negatively to impairment-motivated asset write-downs but positively to restructuring-motivated write-downs. Based on the findings of Francis et al. (1996), we view these restructuring-motivated asset write-downs as an efficiency enhancement consideration by early adopters.[2] More recently, Loh and Tan (2002) [3] report that CEO turnover and macro-economic variables are major determinants of asset write-off decision. Along the same line, Nam and Ronen (2005) suggest that the market actively monitors a migrating CEO’s write-off decision in reckoning the prospects of the destination firm.
The empirical evidence is consistent with our predictions. We find that EARLY firms, pursuing productivity efficiency through restructuring efforts, generate more positive market reaction than the LATE firms at the disclosure of asset write-down. In addition, EARLY firms exhibit higher tax bracket during the period surrounding the event year than LATE firms. The higher effective tax rate for EARLY firms implies that their asset write-down decisions could be partly motivated by the tax consideration; they try to use the asset write-down as a vehicle to reduce the tax burden. Finally, EARLY firms’ asset write-downs are characterized by more capital expenditures, less intangibles, and lower asset turnover ratio and asset age in comparison with LATE firms.
The rest of the paper proceeds as follows. The second section develops hypotheses formulation. The section three describes the sample selection criteria and test methodology. Section four contains the empirical evidence of the paper. Summary and conclusions appear in the final section.
II. Hypotheses Development
There would be little argument that the write-down of capital assets has implications for the future cash flows of the firm. If it were expected, then current stock prices would reflect its effect. If the write-down of capital assets were not anticipated (as we believe is the case with SFAS 121 asset write-down), then prices would react to this new information. A write-down sends a signal that firms’ fixed assets have failed to deliver the performance that was anticipated. A negative association with excess returns suggests that write-down of fixed assets cause the market to revise and lower its expectationsabout the future productivity of the capital assets. Alternatively, the write-down of used assets could be resulted from the rapid changes in technological innovation. With the current pace of changes in new economy it would be hard for them to predict, even in a near-term basis, the timing of the technological innovation. Therefore, firms would write down their assets, not because they necessarily made a poor decision on their asset purchase in the past, but the rapid changes in production environment left them no choice but to write down old assets.
The early forcedretirement of capital assets, however, expect to signal bad news to the market. The basis for this presumption rests on the empirical findings by earlier papers that capital expenditures are good news to the market (Lev and Thiagaranjan, 1993, Kerstein and Kim, 1995, among others). Given this, if a firm has retired a capital asset before its expected useful life is over, it most likely will be considered a signal of failure unless they have legitimate reasons to do so. Since the market had valued the initial investment favorably, it is unlikely that investors were factoring an early retirement of capital assets into their initial calculations. Therefore, there must be a correction to the initial positive reaction.
But there are other prior arguments that imply a positive association with excess returns. A positive association would result if the market perceives the write-down of fixed assets as actions that are necessary to achieve a more productive path in the long run. This would be consistent with John and Ofek (1995) who argue that the market reacts positively to the removal of assets that cause negative synergies.[4] The market would also welcome firms’ action of removing assets which managers believe face obsolescence.
When it comes to voluntary asset write-down, as is the case for EARLY firm, however, this action is more likely to be viewed as management's effort to restructure the organization in order to keep up the pace with technological innovations. To survive the technology-driven competitive environment, EARLY firms (voluntary disclosure firms) may take a proactive measure in order to enhance their production efficiency by writing off the old, under-performing assets before their maturity. Instead of carrying the under-performing assets in their balance sheets, which makes their financial statements and ratios look bad, managers have incentive to replace them with new assets. This would enable them to make investments in cutting-edge technology assets.
Formally stated;
H1: EARLY firms generate more positive market reaction than LATE firms at the disclosure of asset write-down.
These voluntary disclosure firms are also likely to be less efficient in utilizing the existing assets. Since SFAS 121 provides these firms with a legitimate excuse to write down under-performing assets, they can strategically time the adoption of SFAS 121 to their advantage. Therefore, EARLY firms expect to show lower asset turnover rate than LATE firms. Also, a Wall Street Journal article indicates that information technologies are powering U.S. economic growth accounting in recent years for only about 4% of production but for one-third of growth. The demand for tech products is ever-increasing in this new economy, and thus the firms with old, inefficient assets are likely to be the ones who expect to voluntarily write down these assets and replace them with new tech products. EARLY firms, facing the imminent task of restructuring their aging assets, are likely to implement their restructuring efforts by leasing the-state-of-art assets, instead of purchasing them. Therefore, EARLY firms also expect to have greater expenditures for capital leases than for LATE firms.
Assuming the primary motivation of restructuring is to enhance the production efficiency, EARLY firms expect to invest more in the plant assets than in intangibles. Since none of our EARLY firms belongs to the high-tech sector, based on the classification of Francis and Schipper (1999), they are primarily consumers of high-tech products rather than producers of such cutting-edge products.[5] Accordingly, we expect that EARLY firms spend less on R&D and other intangibleacquisitions than LATE firms. Since assets with longer maturity could hinder a firm from handily carrying out the restructuring task more than those with shorter maturity if these assets are unproductive, we expect longer asset age for EARLY firms than for LATE firms. As discussed above, however, EARLY firms with fewer intangible assets than LATE firms, expect to show shorter asset age than for LATE firms as the intangible assets generally have longer useful lives.
Formally stated;
H2: EARLY firms incur more capital expenditures, acquire less intangibles, and exhibit lower asset turnover rate and shorter asset age than LATE firms.
If firms have high net income, they have incentive to write off under-performing assets now, rather than waiting for the mandatory adoption of SFAS 121. This behavior is somewhat consistent with the big bath argument in that firms hedge future earnings against probable future losses due to asset write-downs. Similarly, if EARLY firms are in a high tax-bracket the asset write-down could be used as a vehicle to reduce the tax burden. If EARLY firms’ asset write-down decisions are partly motivated by the existence of sufficient operating cash resources to carry out the restructuring task, we should observe a significant, positive correlation between cash flow from operations and asset write-down of EARLY firms. On the other hand, if EARLY firms belong to a higher tax-bracket, but generate insufficient operating cash for the restructuring efforts, then they expect to be compelled to reduce tax burden through asset write-downs. If this is the case, we expect a negative relation between cash flow from operations and early asset write-down. Therefore, the direction of the operating cash flows in relation to asset write-down decisions is hard to predict, and thus precluded in the third hypothesis.
H3: EARLY firms expect to exhibit higher correlations between asset write-down and both of operating cash flow and tax-bracket than LATE firms.
III. RESEARCH DESIGN
- Sample Selection Criteria
Firms that made voluntary disclosures with respect to SFAS 121 a year before the effective fiscal year of 1996 are drawn from Compact Disclosure. As described in panel A of Table 1, firms must have financial data in COMPUSTAT99 file for the event and previous years and return data in CRSP99 file during the estimation and test period 161 day returns including the 140 day-estimation period and the 3 day event period returns. For the final sample of 47 firms, we searched SEC's EDGAR data base for available 10-K and ARS (Annual Report to Shareholders) filing dates. For 10-K dates, 9 out of 47 firms do not have dates, for which we asked Disclosure Inc. to manually collect the dates for those firms. Since ARS dates are not available from SEC's EDGAR database, we collected them through the help of Disclosure Inc. Two control samples are selected: (i) LATE firms that made disclosures of asset write-down in the mandatory adoption year (1996), but did not make any disclosures in the previous year (the event year: 1995), and (ii) firms in another control sample (MATCH) that are matched with our experimental sample firms in terms of total asset (size) and 4-digit SIC code. Again, Disclosure Inc. collected all ARS dates for both the LATE and MATCH firms and 10-K dates for one third of MATCH firms. To better detect the disclosure effect of asset write-down, we choose either the10-K or the ARS date, whichever is earlier, as the event date.[6]
Panel B of Table 1 shows diversity of the industry composition of both the experimental and control sample. Most of the industries are represented in our experimental sample relatively proportionately, with the exception of the oil and gas extraction industry, which accounts for almost 23.4 % of our sample firms. The second largest industry membership comes from petroleum and coal products industry (12.8%). LATE firms are also represented by various industries, at least as evenly as EARLY firms. The two most conspicuous industries belonging to LATE firms are eating and drinking places (15.2 %), and electric, gas, and sanitary (12.1%) industries, respectively.
[Insert Table 1]
- Test Methodology for Stock Market Reaction
Following Patell's (1976) test procedures, standardized prediction error (SPE) is computed using the market model of Sharpe (1964) and Lintner (1965) as follows:
During the estimation period,
Ri = i + i * Rm + ei (1)
where: Ri = return on security i at time ,
Rm= return on market portfolio at time (CRSP value-weighted index).
i = E(Ri) - i * E(Rm),
i = Cov(Ri,Rm)/Var(Rm)
ei = error term
i = 1,....., N, firm index, and
= 1,....., T, time index.
During the three-day event period,