«Leif Atle Beisland University of Agder Dissertation submitted to the Department of Accounting, Auditing and Law at the Norwegian School of Economics ...»
One explanation is that earnings reflects cash flow forecasts and has a higher correlation with value then does current cash flow. Earnings’ inclusion of those forecasts causes earnings to be a better forecast of (and so a better proxy for) future cash flows than current cash flows (Dechow et al., 1998, pp. 133-134).
Since the results of prior research document deterioration in the relation between prices and earnings over time, it is puzzling that the relation between earnings and future cash flows is generally increasing over time, because stock price should equal the present value of future cash flows (M. Kim & Kross, 2005, p. 775).
A fundamental question in accounting is the relative ability of accrual-based earnings and cash flows to predict a firm’s ability to generate future cash flows (Subramanyam & Venkatachalam, 2007, p. 457).
definition is not debated much in the above listed articles. However, some arguments for the
choice are put forward:
Since the primary interest is in operating performance measures and assessing what information, if any, about future cash flows is provided by accruals in the income determination process, the focus in this paper is on operating cash flows rather than total cash flows (Rayburn, 1986, p. 114).
Cash flow from operations reflects the net cash flows generated by the firm’s operating activities. This measure includes accruals that are long-term in nature (i.e., do not reverse within one year) and mitigate timing and matching problems associated with the firm’s investment and financing activities (Dechow, 1994, p. 8).
The only article that I have found that actually discusses the use of cash flow from operations versus free cash flow is Subramanyam and Venkatachalam (2007). Even though they decide to investigate cash flow from operations in their empirical study, they state that “…operating cash flows are not value attributes. This is because operating cash flows ignore investments in operating assets; the appropriate value attribute is free cash flows” (Subramanyam & Venkatachalam, 2007, p. 461). They state that the reason why they still apply cash flow from operations is that this variable has been used traditionally in the literature to evaluate value relevance.
Cash flow from operations is sometimes viewed as a better short term measure of firm performance than free cash flow from operations. Investments decrease the current free cash flow. Thus, investments may be implicitly regarded as “negative” if the time horizon is too short. Free cash flow may give an incorrect picture of the cash flow generating capabilities of companies that temporarily have large investment expenditures, for instance, companies that are early in their life-cycle. Dechow (1994) claims that the choice of cash flow concept
Still, company value is undoubtedly a function of cash flow not only from historical investments but also from current investments and from known or expected future investments. Rayburn (1986) states that her choice of cash flow definition is due to her focus on information content of accruals. When disregarding capital expenditures in the cash flow concept, the difference between net income and cash flow is equal to the accruals. Accruals are definitely of particular interest in accounting research. One can argue that the choice of accounting framework, i.e., accounting law and standards, is materialised through the accruals. Hence, it may be natural to disregard the “objective” investment expenditures31. It should also be noted that the cash flow definition applied in the quoted research is equal to free cash flow to equity if one assumes that capital expenditures are financed through net financial assets32. Although such an assumption probably is highly unreasonable, it would make the analyses consistent with the original cash flow valuation model. Some databases lack information on capital expenditure and make this assumption tempting to use. It can also Dechow (1994) analyses the value relevance of both cash flow and earnings. One may claim that if investment decisions are disregarded in the cash flow definition, one should disregard the financial items, i.e., use operating earnings, in the earnings definition. The cash flow and earnings definitions may then be more comparable.
However, Dechow (1994) applies an earnings definition inclusive of financial items (but exclusive of extraordinary items).
Although the classification of cash flow might be highly subjective.
Assume no dirty surplus. Then:
Free Cash Flow to Equity = Earnings - Change in Equity = Earnings - (Change in Net Operating Assets - Change in Net Financial Debt) = Earnings - (Capital Expenditures - Depreciation & Amortisation + Change in Working Capital) + Change in Net Financial Debt = Earnings + Depreciation and Amortisation – Change in Working Capital - (Capital Expenditures – Change in Net Financial Debt) If Capital Expenditures = Change in Net Financial Debt (i.e., investments are financed through net financial debt), then Free Cash Flow from Equity = Earnings + Depreciation and Amortisation – Change in Working Capital
that is necessary to maintain current operations33. Nevertheless, none of the listed papers use any of these lines of reasoning to substantiate their cash flow definition. Note that the FASB, in its assertions that earnings are a better cash flow predictor than current cash flow, does not specify to which cash flow concept they are referring. One cannot disregard the possibility that Kim and Kross (2005) (see section 2) are unable to reconcile the increase in earnings’ ability to predict cash flows with earnings’ decreasing association with stock return simply because they apply cash from operations in their empirical study.
The intention with this discussion is to question the choice of cash flow definition used in this research tradition. Since cash flow is viewed as a value driver, it may be peculiar to use a cash flow concept that is not the input in any valuation model. There may be good arguments for the choice, but most papers tend to ignore such a discussion.
Feltham and Ohlson (1996) develop a valuation model where company value equals the present value of future expected cash receipts due to prior and current periods’ investments plus the present value of future expected investments (both replacements and expansions).
AbstractIt has been claimed that accounting information is losing relevance over time as firms increasingly rely on resources that cannot be capitalized under the existing accounting regime.
We find that easily detected transitory elements of reported earnings cause visible differences in value relevance between traditional and non-traditional industries. When the different properties of earnings components are considered, non-traditional industries provide no less relevant information to the market. We furthermore extend past research by investigating changes in value relevance over time and find that the non-traditional industries experience considerably more variance, independent of the transitory elements of reported earnings. This variance is associated to both stock market sentiments and the growth of the economy. When the economy is doing well and firms are highly valued in relation to fundamentals, accounting information is considerably less able to explain security prices in non-traditional industries.
* We would like to thank Jiri Novak for his suggestions and comments on earlier versions of this paper.
information plays an important role in the investors’ forecasts of these cash flows. A considerable amount of research has examined if the value relevance of accounting information changes over time (e.g., Collins, Maydew, and Weiss , Brown, Kin, and Lys , Francis and Schipper , Lev and Zarowin , Aboody, Hughes, and Liu , Dontoh, Radhakrishnan, and Ronen , Goodwin and Ahmed ). A number of these studies find that accounting information over time has indeed become less relevant for investors (e.g., Brown, Kin, and Lys , Lev and Zarowin ). A common explanation is that the accounting system, based on conservative accounting principles, fails to reflect the situation of today’s enterprises that increasingly rely on resources which cannot be recognized (e.g., Lev and Zarowin , Goodwin and Ahmed ). In particular it appears as if the reported earnings has lost relevance over time (Collins, Maydew, and Weiss , Brown, Kin, and Lys , Lev and Zarowin ). This causes great concern for timing and matching problems and the extent to which the accounting system is able to allocate cash flows from irrelevant periods of time, to the relevant periods.
We use a data set consisting of essentially all firms listed at the Stockholm Stock Exchange in the years 1979 to 2004. The firms are divided into two broad industry sectors; traditional and non-traditional industries. While several of the past studies have identified changes in value relevance over time (usually seen as trends) none of them have closely examined intertemporal differences in value relevance across industries. We do so, and with a particular focus on transitory components of earnings. Overall, it is expected that the nature of nontraditional industries creates a greater variance in the level of association between accounting information and security prices/returns. However, if we are able to detect the causes for crosssectional differences in value relevance, then it is likely that investors can do likewise.
ability of explaining security prices/returns in non-traditional industries as to traditional industries. Because of somewhat inconclusive past findings we expect no difference in value relevance between these two groups on average as any temporary difference cancel out over a prolonged period of time. While past studies only take nonlinearities in the form of losses into account, we also consider variations in positive earnings. Also profits contain a transitory element and taking the different properties of sustainable and transitory earnings into account, both the overall value relevance and the relative value relevance of non-traditional industries increase. In addition we find no decrease in value relevance over time and return model specifications consistently suggest that the accounting information is more value relevant in the non-traditional industries.
The second part of the paper concerns inter-temporal variations in the overall explanatory power of accounting information in traditional and non-traditional industries. We expect nontraditional industries to experience more time-varying value relevance. There are several reasons for this: If the firm value to a greater extent is determined by resources that cannot be capitalized then profits might be lower in good times (because of investments that are expensed) and higher in bad times. The non-traditional industries might also be more uncertain than traditional industries and the uncertainty can be leveraged by market sentiments. An example of this is the so-called IT-bubble in the late 1990s when many nontraditional firms were expected to experience a very rosy future. The relative value relevance of accounting information for the non-traditional industries in this time period was exceptionally low. We expect an association between the relative explanatory power of accounting information provided by firms operating in non-traditional industries, and the state
information is less relevant in periods of time when investors have high expectations for the future. Such a loss of relevance affects firms in non-traditional industries more than those operating in traditional industries. Also these expectations are confirmed. We find a higher variance in the value relevance within the non-traditional industries. We argue that such variations, in a long-term perspective, also constitute a measure of value relevance. The variations for non-traditional industries appear to be explained by macro-economic factors and market sentiments.
The remainder of the paper is organized as follows. Section two contains a theoretical background and research hypotheses. Section three deals with the research design and describes the data. Section four contains the empirical analysis and the fifth section concludes.
2 Theoretical Background A number of different definitions of value relevance exist in the accounting research literature (e.g., Francis and Schipper ), and these are not necessarily mutually exclusive. We define value relevance as the ability of financial statement information to capture and summarize information that determines the firm’s value. Thus, value relevance is measured as the degree of statistical association between accounting information and market values or returns over a long time horizon.
During the last decade there has been an interest in the long term development of accounting information’s relevance to investors. Most of this research has been conducted in the U.S.
where there is rather undisputed evidence that accounting information has lost some of its relevance over time (Collins, Maydew, and Weiss , Chang , Brown, Kin, and
The decrease in value relevance seems to be related to a lower relevance of earnings (e.g., Collins, Maydew, and Weiss , Lev and Zarowin ). While Collins, Maydew, and Weiss  find that the loss in value relevance of accounting earnings is compensated by an increase in the relevance of book value of equity, other studies suggest that the decrease in the relevance of accounting earnings is the main reason why the overall relevance has gone down (e.g., Lev and Zarowin, 1999).