استهلاک و ارزش بازار درآمد

The purpose of this paper is to better understand how financial markets use depreciation information that appears in the financial statements. When computing depreciation expense, a firm has discretion in its choice of depreciation method and this choice can impact the amount of depreciation expense and therefore reported earnings. Ceteris paribus, we claim that a firm which reports a low depreciation ratio (i.e. the ratio of depreciation expense to average gross property, plant, and equipment) tends, on average, to report understated depreciation and therefore overstated earnings, relative to similar firms that report moderate or high depreciation ratios. We expect market participants to adjust such a firm's overstated earnings downward. Accordingly, we hypothesize that firms with low depreciation ratios tend to have low earnings response coefficients (ERCs), where the ERC is the coefficient obtained from regressing a price metric on an earnings metric. We find significant empirical evidence to support this hypothesis. The results hold while controlling for factors such as risk, expected growth, size, capital intensity, and industry. The results also do not appear to be attributable to differences in earnings persistence. The evidence suggests that financial markets are not fixated on reported earnings; instead, markets are making adjustments for the various depreciation methods selected by firms.

When analyzing financial statements and drawing comparisons across firms, financial analysts can encounter situations where firms differ in how they elect to report depreciation . For example, firms can differ in their choice of useful asset life . Generally accepted accounting principles allow firms to elect to depreciate their fixed assets, within reason, over relatively long or short asset lives . Generally, firms electing long asset lives for depreciation will report, ceteris paribus, lower depreciation expense and higher earnings than similar firms using short asset lives . There are of course many other choices that affect depreciation (e .g . choice of salvage value) . For our purposes, the key pattern is that depreciation choices can have a bearing on the level of depreciation expense and therefore reported earnings . Unlike prior research (e .g. Holthausen, 1981), this paper does not try to test differences based on the firm's choice of depreciation method . Rather we focus on a depreciation ratio (depreciation expense/average gross property, plant, and equipment) that serves as a composite proxy for depreciation choices made by a firm. After controlling for industry and other factors, we claim that firms with low depreciation ratios report relatively low depreciation expense and have overstated earnings .This paper is motivated by a desire to better understand how financial markets use depreciation information that appears in financial statements . Prior research discusses how depreciation can be important to investment professionals . Siegel (1982) surveys accountants, security analysts, and financial managers . These professionals report that depreciation is an important issue bearing on the analysis of a firm's earnings . They also feel that certain depreciation methods can result in overstated earnings . Based on the survey results, Siegel (1982 : 64) claims, "Furthermore, a firm's failure to provide adequate depreciation charges for the possible obsolescence of plant assets results in overstated earnings . . ." Our paper follows this reasoning and conducts empirical tests on market adjustments for depreciation choices. This paper also extends the substantial literature that began with Ball and Brown's (1968) seminal study on the relationship between earnings and stock prices . Recently, this line of literature has shown increased interest in the crosssectional variation in the earnings response coefficient (ERC), the coefficient obtained from regressing a price metric on an earnings metric. These studies focus on the properties of the earnings-generation process, such as risk, growth, and persistence e .g. (Kormendi and Lipe (1987), Easton and Zmijewski (1989), Collins and Kothari (1989), Biddle and Seow (1991), and Ali and Zarowin (1992a, b)) . Lev (1989, p . 172) notes, however, that this type of research is notparticularly revealing nor will it provide new, deeper insights into how earnings are used by investors . In fact, Lev (1989) suggests that relatively little has been learned about the usefulness of earnings in the two decades following Ball and Brown (1968) and calls for researchers to explore how the market uses earnings information . Our paper responds to Lev's call by studying whether the market appears to use earnings only after adjusting for cross-sectional differences in depreciation choices . We find significant evidence indicating that firms with low depreciation ratios tend to have lower ERCs . This evidence is consistent with a conclusion that firms with low depreciation ratios tend to report low depreciation expense and therefore overstate earnings, and the market adjusts these earnings downward . The results hold while controlling for factors such as risk, expected growth, size, capital intensity, and industry . The results also do not appear to be related to differences in earnings persistence . The evidence suggests financial markets are not fixated on earnings ; instead, markets are adjusting the valuation of earnings for the various depreciation choices made by firms . The remainder of this paper is organized as follows . Section 2 formally develops our hypothesis. Section 3 describes the sample and the variables . Section 4 provides the empirical analysis and results . Section 5 offers a summary and conclusion .

This study examines the relationship between depreciation ratios and ERCs . We claim that firms with low depreciation ratios tend, on average, to understate depreciation and therefore overstate earnings . We expect the market to adjust the overstated earnings downward, resulting in a positive relationship between depreciation ratios and ERCs . In other words, firms with low depreciation ratios are expected to have overstated earnings and therefore low ERCs . As predicted, the results show a positive relation between ERCs and depreciation ratios . The positive relation holds after controlling for other factors that have been shown to be cross-sectionally related to ERCs (e .g . risk, expected growth, size, and capital intensity) . The results do not seem to be driven by differences in earnings persistence . The earnings persistence of the high and low deviation firms is nearly identical . The results also hold within most industries indicating that the results are not driven by differences in industry factors . Most importantly, when the effects of depreciation are eliminated from earnings, no evidence is found for the relationship between ERCs and depreciation ratios . This suggests that depreciation alone is likely driving the smaller ERCs for firms with low depreciation ratios . The evidence suggests financial markets are not fixated on reported earnings . Instead, markets are making adjustments for the various depreciation methodsselected by firms . These findings provide empirical evidence in support of theassertions made in Siegel (1982) ; financial analysts appear to be adjusting their evaluations of a firm's earnings based on depreciation information appearing in the financial statements . Financial statement users should consider differences in depreciation ratios when making cross-sectional comparisons of reported earnings .