1
Accounting Conservatism and the Relation Between Returns and Accounting Data
Peter Easton
The Ohio State University and The University of Melbourne
and
Jinhan Pae
Queen’s University
May 2003
We thank Ashiq Ali, Bruce Billings, Keji Chen, Dan Dhaliwal, Jerry Feltham, Rick Morton, Jim Ohlson, Gord Richardson, Greg Sommers, Mark Trombley, Bill Schwartz, Philip Stocken, two anonymous referees, and workshop participants at the University of British Columbia, Florida State University, Nyenrode University, the Ohio State University, the University of Arizona, and the 2001 American Accounting Association annual meetings, for helpful comments on earlier drafts.
1
Accounting Conservatism and the Relation Between Returns and Accounting Data
Abstract
The study adds change in cash investments and change in lagged operating assets to the regression of returns on earnings levels and earnings changes examined in Easton and Harris (1991). The model in Feltham and Ohlson (1996) suggests that a positive coefficient on change in cash investments captures conservatism associated with investments in positive net present value projects the effects of which will not flow into the accounting statements until the expected future benefits are realized. A positive coefficient on change in lagged operating assets implies accounting conservatism associated with the application of accounting rules (such as expensing rather than capitalizing R&D expenditures) to operating assets in place. Our empirical results are, in general, consistent with the predictions from Feltham and Ohlson (1996). We examine differences in conservatism across samples with different market to book ratios, we compare firms with non-negative returns with firms with negative returns, we compare firms reporting losses with firms reporting profits, and we examine firms in different industries, firms with different levels of research and development expenditure, different amounts of depreciation, different amounts of advertising expense, and firms that adopt LIFO inventory valuation compared with those that adopt an alternative to LIFO.
1
1. Introduction
The study adds change in cash investments and change in lagged operating assets to the regression of returns on earnings levels and earnings changes examined in Easton and Harris (1991).[1] We show that change in cash investments and change in lagged operating assets have incremental explanatory power (over earnings levels and earnings changes) for returns suggesting that accounting is conservative in the Feltham and Ohlson (1996) sense.
The model in Feltham and Ohlson (1996) suggests that a positive coefficient on change in lagged operating assets implies accounting conservatism associated with the application of accounting rules (such as expensing rather than capitalizing R&D expenditures) to operating assets in place. A positive coefficient on change in cash investments captures conservatism associated with investments in positive net present value projects the effects of which will not flow into the accounting statements until the expected future benefits are realized. Our empirical results are, in general, consistent with the predictions from Feltham and Ohlson (1996) providing evidence of accounting conservatism.
Feltham and Ohlson (1995, 1996) generalize and extend Ohlson (1995) by introducing the notion of accounting conservatism (defined as market value exceeding book value in the long run). With few exceptions empirical studies of the relation between returns and accounting data rely on Ohlson (1995) implicitly ignoring the effects of conservatism. Easton and Harris (1991) show that deflated (by beginning-of-period stock price) earnings levels and deflated earnings changes have significant explanatory power for annual returns. Their regression is implicitly based on Ohlson (1995) in which accounting is assumed to be unbiased (that is, the market value of equity equals the book value of equity in the long run). However, the current disparity between market value of equity and book value of equity suggests that accounting is not likely to be unbiased.
Empirical tests are performed on a sample of 53,554 firm-year observations from 1988 to 2001. The mean of the year-by-year estimates of the coefficients on change in lagged operating assets is not significantly different from zero while the mean of the estimate of the coefficient on change in cash investments is 0.266 (t-statistic of 7.56). The significantly positive estimate of the coefficient on change in cash investments suggests that accounting is conservative, consistent with the results in Ahmed, Morton, and Schaefer (2000).
We find that accounting conservatism as defined in Feltham and Ohlson (1996) is more pronounced for firms where the market value of operating assets is high relative to their book value (this is consistent with use of the ratio of the market value of common equity to the book value of common equity as a proxy for accounting conservatism in Beaver and Ryan (2000)).[2] Our empirical analyses demonstrate that the magnitude of the estimates of the coefficients on change in cash investments and change in lagged operating assets increase as the price to book ratio increases. Industry-specific analysis shows evidence of varying degrees of accounting conservatism across industries. Accounting conservatism is particularly evident in the pharmaceutical industry.
Following Basu (1997) who suggests that accounting will be more (less) conservative for firms with good (bad) news over the fiscal period,we seek evidence of a difference in accounting conservatism between firms with non-negative returns (good news) and firms with negative returns (bad news). Consistent with Basu (1997), the estimate of the coefficient on change in cash investments is significantly less for firms with net bad news (negative returns) than the estimate of the coefficient on change in cash investments for firms with net good news (non-negative returns). The estimate of the coefficient on change in lagged operating assets is significantly negative for firms with negative returns while it is not significantly different from zero for firms with non-negative returns. In the Feltham and Ohlson (1995, 1996) model lagged operating assets have negative weight if there is under-depreciation of these assets. That is, for these firms where the market assessed the net change in value as negative, accounting recorded a smaller decline in value. The negative estimate of this coefficient suggests that the accruals of firms with net bad news are not sufficient to make accounting earnings fully reflect the decline of firm value in years of negative returns.
Hayn (1995) focuses on the news in earnings rather than the news in returns to motivate an analysis of the returns/earnings relation for firms reporting losses compared with firms reporting profits. In order to examine the effects of losses on accounting conservatism, we partition the sample into profit and loss firms. Our empirical analyses suggest that there is no difference in accounting conservatism associated with the application of accounting rules between firms reporting losses and firms reporting profits. The estimates of the coefficients on change in cash investments are positive for both profit and loss firms consistent with the notion that cash investments are generally in positive net present value projects.
Several studies have examined whether accounting is on averageconservative. Even though the generally positive difference between market value and book value of equity suggests that accounting is likely to be conservative, the extant empirical studies based on Feltham and Ohlson (1995, 1996) do not provide convincing evidence. Accounting conservatism is usually examined by estimating either the Feltham and Ohlson (1995) information dynamics or the implied valuation function.[3] Contrary to the general belief that accounting is conservative, empirical studies based on the information dynamics generally provide evidence that is consistent with aggressive accounting (Dechow, Hutton, and Sloan 1999; Myers 1999; Ahmed, Morton, and Schaefer 2000). The results of empirical studies of the valuation relation are mixed. Myers (1999) provides empirical evidence consistent with unbiased or aggressive accounting while Ahmed, Morton, and Schaefer (2000) report empirical results consistent with conservative accounting.[4]
In this paper, we seek evidence of conservatism using the valuation relation rather than the information dynamics.[5] We provide empirical evidence of accounting conservatism using the return specification rather than the levels specification: the empirical results based on the return specification are less susceptible to scale effects, omitted variables, and multicollinearity. We show that the degree of accounting conservatism varies with the ratio of market value to book value and it differs across industries, we demonstrate the different effects of accounting conservatism associated with accounting rules and accounting conservatism associated with investment in positive net present value projects, and we show some consistency between the empirical implications gleaned from the Feltham and Ohlson (1995, 1996) notion of conservatism and other notions of conservatism in the extant literature.
The paper proceeds as follows. In section 2, we develop an empirical regression model that is implied by Feltham and Ohlson (1996). Section 3 describes the data and the sample selection procedure. Section 4 reports the results of the empirical analysis. We summarize these results and draw conclusions in section 5.
2. Model
Our empirical regression model is based on Feltham and Ohlson (1996) rather than Ohlson (1995). In later analyses we consider the empirical implications of Feltham and Ohlson (1995). Feltham and Ohlson (1996) explicitly model the possibility of conservative accounting whereas Ohlson (1995) implicitly assumes unbiased accounting. As in Feltham and Ohlson (1996), accounting is defined as conservative (unbiased) if the value of a firm is expected to exceed (equal) the firm's book value of equity in the long run.
The basic ingredients of the Feltham and Ohlson (1996) model are:
(1) the cash flow dynamics:[6]
(CFD)
(2) the present value of cash flow relation:
(PVCF)
and (3) the operating asset relation:
(OAR)
where > 0, [0,1), [0, R), crt and cit are cash receipts and cash investments, ct( crt - cit) is net cash flows, Vt is the market value of operating assets at date t, R is one plus the riskless rate of return (r), oat is net operating asset at time t, oxt is operating earnings, and oact is operating accruals for period t.
Assumptions CFD, PVCF, and OAR are sufficient to derive the following valuation relation:
(1)
where, and is residual operating income.
Invoking the Feltham and Ohlson (1995) assumption that financial assets are valued at market, equation (1) may be re-written as:
(2)
where ptis the market value of equity at time t, pt is the book value of equity at time t and xta is residual earnings for period t (that is, xt – (R-1)bt-1).[7]
Feltham and Ohlson (1996) examine the effect of accounting accruals on the relation between accounting numbers and firm value by focusing on the relation between operating accruals (the accounting measure of change in value of operating assets – referred to generically as depreciation) and beginning of period book value of operating assets. In the interest of parsimony, they invoke the simplifying assumption that operating accruals are a constant proportion (1-) of beginning of period operating assets. Although this proportion (1-) may be viewed as an accrual that has the characteristics of the declining balance depreciation method, it will also capture any change in the accounting measure of the value of operating assets. For example, research and development costs that are expensed immediately have a of zero. In general, may be viewed as the weighted average of the accounting measures of change in value. Under this simplifying assumption, Feltham and Ohlson (1996) suggest the following value relation:
(3)
where .
Taking first differences and rearranging yields:
(4)
The coefficient on change in lagged operating assets captures the effect of accounting value added, and the coefficient on change in cash investments captures the effect of economic value added. We elaborate on this point.
Recall that . That is, change in lagged operating assets (oat-1) has positive weight if there is over-depreciation of these assets (that is, the accounting depreciation rate (1-) is greater than the economic depreciation rate (1-)).[8] In other words, if assets have been over-depreciated in the past, future earnings will be higher because there will be less depreciation to match against revenues. Since this higher earnings is due to the accounting policy (depreciation rate), the consequent positive residual income reflects only accounting value added – not economic value added. If , operating assets are recorded at market value – and thus goodwill (the present value of future residual income) associated with depreciation of beginning operating assets is zero.
Recall that . That is, current investments cit have positive weight if they have positive NPV (that is, > 1). Note that cash investments at the end of the current period do not affect residual income of the current period (this captures the reality that new investments do not affect accounting earnings until products are sold or a service is provided). The effect of current investments is on residual income of future periods. That is, positive residual income in future periods may reflect economic value added in past periods but positive residual income in the current period does not reflect economic value added in the current period.
To summarize, if accounting is conservative as in Feltham and Ohlson (1995, 1996), change in cash investments, change in lagged operating assets, and lagged dividends should be included as additional variables in the return, earnings levels, and earnings changes regression as in equation (4). The lack of consideration of accounting conservatism may affect the estimates of the coefficients on earnings levels and earnings changes because change in cash investments, change in lagged operating assets and/or lagged dividends may be correlated with these variables.
Most of our analyses are based on regressions that are the empirical analogue of equation (4) with all variables scaled by beginning market value of equity: [9]
(5)
where .[10] The subscript j denotes an observation for firm j. The cross-sectional regressions are run separately for each year of available data. captures the effect of conservatism due to future positive NPV projects, and captures the effect of conservatism due to accounting rules. The coefficients on earnings levels, earnings changes, and lagged dividends are all predicted to be positive.[11]
3. Data Selection and Sample Description
Initially, we collect all Compustat firm-year observations from fiscal years 1988 through 2001 for which we have complete data for the following items. Return () is obtained from CRSP by compounding monthly returns during the fiscal period. We use comprehensive income as a measure of earnings (). Comprehensive income is net income (#172) minus preferred dividends (#19) plus the change in value of marketable securities (#238) plus the change in cumulative foreign currency translation adjustment (#230). The change in comprehensive income is denoted by xt. Dividends () are the sum of dividends to common shareholders (item #21) and net capital contributions. Net capital contributions are purchases of common and preferred stock (item#115) minus sales of common and preferred stock (item#108). Operating assets () are book value of equity () minus financial assets (). Book value of equity (bt) is common equity (#60) plus preferred treasury stock (#227) minus preferred dividends in arrears (#242). Financial assets (fat) are cash and short-term investments (#1) plus investments and advances-others (#32) minus debt in current liabilities (#34) minus long-term debt (#9) minus preferred stock (#130) plus preferred treasury stock (#227) minus preferred dividends in arrears (#242) minus minority interest (#38). Cash investments are obtained from the cash flow statement as the negative of cash flows from investing activities (#311).[12]
The ratio of the market value of operating assets to the book value of operating assets (V/oa) is the market value of common equity minus financial assets (fat) divided by the book value of operating assets ((pt - fat)/oat). All variables except the market value of equity (), annual stock returns (), and ratio of the market value of operating assets to the book value of operating assets (V/oa) are deflated by the beginning market value of equity (). Observations with negative book value of equity or negative book value or market value of operating assets are excluded. We further delete observations in the top and bottom one percent of the distribution for any one of the following variables: annual returns, earnings levels, earnings changes, lagged dividends, change in cash investments, and change in lagged operating assets in order to mitigate the effect of extreme values.
The final sample is 53,554 firm-year observations, which consist of 37,169 firm-year observations reporting profits (profit firms, hereafter) and 16,385 firm-year observations reporting losses (loss firms, hereafter) and 28,366 observations with non-negative returns (good news firms, hereafter) and 25,188 observations with negative returns (bad news firms, hereafter). The lack of data necessary to measure cash investments (cit) restricts most of our analysis to the post-1987 period.
Panel A of table 1 reports descriptive statistics for the sample of 53,554 firm-year observations from 1988 through 2001. The median market value of equity is $128.26 million. Over the 14 years, the mean and median annual raw stock returns are 12.2% and 3.2%, respectively. Median net comprehensive income and the change in net comprehensive income are 4.3% and 0.4% of the beginning market value of equity. Median lagged dividends are zero. The decomposition of book value of equity into operating assets and financial assets shows that firms have on average net financial obligations; hence operating assets are greater than book value of equity. The positive change in operating assets (median of 3.8 percent of price) implies that operating assets are on average increasing. The ratio of the market value of operating assets to the book value of operating assets is generally greater than one although, for about 18 percent of the sample, market value of operating assets is less than their book value.
Panel B of table 1 reports descriptive statistics for profit and loss sub-samples. Profit firms are, on average, bigger than loss firms. The median market values of equity for profit and loss firms are $214.68 million and $43.51 million, respectively. Loss firms have, on average, higher market to book (P/B and V/oa) ratios than profit firms. This is due to both higher market value of equity for profit firms and lower book value of equity for loss firms.