The information content of economic value-added: A comparative analysis with earnings, cash flow and residual income

Transcription

1 The information content of economic value-added: A comparative analysis with earnings, cash flow and residual income Tracey West a, Andrew Worthington b,* a School of Accounting, Banking and Finance, Griffith University, Nathan, QLD 4111, Australia b School of Economics and Finance, Queensland University of Technology, Brisbane, QLD 4001, Australia Abstract Pooled time-series, cross-sectional data on 110 Australian companies over the period is employed to examine whether the trademarked variant of residual income known as economic value-added (EVA ) is more highly associated with stock returns than more conventional accounting-based measures. These other measures of internal and external performance include earnings, net cash flow and residual income. Relative information content tests reveal returns to be more closely associated with earnings than net cash flow, residual income and EVA respectively However, consistent with the construction of EVA, incremental information content tests suggest that EVA adds more explanatory power to earnings than either net cash flow or residual income. An analysis of the components of EVA confirms that the capital charges and GAAP-related adjustments most closely associated with EVA are significant at the margin in explaining market returns. JEL classifications: C23; G14; M41. Keywords: Value-relevance; relative and incremental information content; economic-value added; residual income. 1. Introduction An accepted financial axiom is that the role of managers is to maximise the wealth of shareholders by the efficient allocation of resources. In order to operationalise this objective, shareholder wealth is traditionally proxied by either standard accounting magnitudes (such as profits, earnings and cash flow from operations) or financial statement ratios (including earnings per share and the returns on assets, investment and equity). Managers, shareholders and other interested parties then use this financial statement information to assess and predict current and future performance. Furthermore, under the semi-strong form of the efficient market hypothesis, the publicly available information contained in these variables is posited to be readily interpreted by the market and thereby incorporated into future stock prices. Unfortunately, the empirical literature to date suggests that there is no single accounting-based measure upon which one can rely to explain changes in shareholder wealth. This is despite the fact that such a measure would prove * Corresponding author. Tel: 61 (0) ; Fax: 61 (0) ; a.worthington@qut.edu.au

2 2 invaluable to the various parties interested in aspects of firm performance. Lee (1996), for example, argues that the search for a superior measure of firm valuation is a key feature of contemporary empirical finance: For years, investors and corporate managers have been seeking a timely and reliable measurement of shareholders wealth. With such a measure, investors could spot over- or underpriced stocks, lenders could gauge the security of their loans and managers could monitor the profitability of their factories, divisions and firms. One professedly recent innovation in the field of internal and external performance measurement is a trademarked variant of residual income (net operating profits less a charge for the opportunity cost of invested capital) known as economic value-added (EVA ). As a starting point its developer and principal advocate, US-based business consultants Stern Stewart, argue that earnings, earnings per share, and earnings growth are misleading measures of corporate performance [and that] the best practical periodic performance measure is economic value-added (Stewart, 1991, p. 66). In addition, EVA is the financial performance measure that comes closer than any other to capturing the true economic profit of an enterprise. EVA also is the performance measure most directly linked to the creation of shareholder wealth over time (Stewart, 1991, p. 66). Citing inhouse research, Stewart (1994, p. 75) also suggests that EVA stands well out from the crowd as the single best measure of wealth creation on a contemporaneous basis [and] is almost 50% better than its closest accounting-based competitor [including EPS, ROE and ROI] in explaining changes in shareholder wealth. Support for EVA has also been forthcoming from other sources. Fortune has called it today s hottest financial idea, The Real Key to Creating Wealth (30 September 1993) and a A New Way to Find Bargains (9 December 1996), and Peter Drucker in the Harvard Business Review suggested that EVA s growing popularity reflected the demands of the information age for a measure of total factor productivity (Stern, Stewart and Chew, 1995, p. 32). McClenahen (1998) similarly observes that traditional corporate performance measures are being relegated to second-class status as metrics such as EVA become management s primary tools. Finally, there has been the widespread adoption of EVA by security analysts since instead of using a dividend discount approach, these models measure value from the point of view of the firms capacity for ongoing wealth creation rather than simply wealth distribution (Herzberg, 1998, p. 45) (emphasis added).

3 3 In response to these claims, an emerging literature has addressed the empirical issue as to whether EVA is more highly associated with stock returns and firm values than other less prosaic accounting figures. For example, Biddle et al. (1997) used relative and incremental information tests to examine whether stock returns were more highly associated with EVA, residual income or cash flow from operations. Biddle et al. (1997, p. 333) concluded that while for some firms EVA may be an effective tool for internal decision making, performance measurement, and incentive compensation, it does not dominate earnings in its association with stock market returns. Chen and Dodd (1997) likewise examined different dimensions of the EVA system and concluded: not a single EVA measure [annualised EVA return, average EVA per share, change in standardised EVA and average return on capital] was able to account for more than 26 percent of the variation in stock return. Lehn and Makhija (1997) Rogerson (1997) and Biddle, Bowen and Wallace (1997) reached similar conclusions. Lastly, Clinton and Chen (1998) compared share prices and returns to residual cash flow, economic value-added and other traditional measures, and recommended that companies using EVA consider residual cash flow as an alternative. However, Bao and Bao (1998, p. 262) in an analysis of price levels and firm valuations concluded that the results are not consistent for earnings and abnormal economic earnings, but are consistent for value added, i.e., value added is significant in both levels and changes deflated by price analyses. Similarly, Uyemura, Kantor and Petit (1996) demonstrated that EVA has a high correlation with market value added (the difference between the firm s value and cumulative investor capital) and thereby stock price, while O Byrne (1996) estimated that changes in EVA explain more variation in long-term stock returns than changes in earnings. Finally, and from a stock selection perspective, Herzberg (1998, p. 52) concluded that the residual income valuation model (including EVA) appears to have been very effective in uncovering firms whose stock is underpriced when considered in conjunction with expectations for strong earnings and growth. Nevertheless, the bulk of empirical evidence indicates that the superiority of EVA vis-à-vis earnings (as variously defined) has not been forthcoming. When examining existing research in this area, a number of salient points emerge. First, notwithstanding that EVA figures are readily available and promoted in the UK, Australia, Canada, Brazil, Germany, Mexico, Turkey and France [see Stern Stewart (1999)], no empirical studies of this type (as far as the authors are aware) have been conducted outside the United States. This is despite several international companies adopting EVA for performance measure and/or

4 4 incentive compensation packages. For example, in Australia the ANZ Banking Group, Fletcher Challenge Limited, James Hardie Industries and the Wrightson Group, amongst others, have implemented EVA financial management systems (Rennie, 1997). There is an obvious requirement to examine the usefulness of EVA vis-à-vis traditional financial statement measures in an alternative institutional milieu. Second, Biddle et al. (1997, p. 303) suggest that data on the information content of EVA [and other measures of performance] provide potentially useful input to the normative policy debate on what performance measures(s) should be reported in financial statements. In addition, in Australia as elsewhere, there is renewed interest in the types of information provided by investment analysts, with EVA studies now appearing in stockbroking reports (Wood 1996). Both developments are equally deserving of empirical attention. Third, Biddle et al. (1997, p. 333) in a study of the usefulness of EVA in the US context suggest that an avenue for future research suggested by the findings of this study is to examine more closely which components of EVA and earnings contribute to, or subtract from, information content. Put differently, given that EVA consists of nearly 160 potential changes to accounting figures grouped across adjustments to accounting measures of operating profits and capital, there is the requirement to quantify the contribution of these sub-components to overall firm performance. Finally, there has been an emphasis in previous empirical work in this area on either single crosssections of listed companies or limited pooled time-series, cross-sectional data. For example, Bao and Bao (1998) employ a cross-section of 166 firms over the period 1992/93. Contrary to the oftmaintained hypothesis of semi-strong form market efficiency, it may be the case that market participants only recognise the reporting benefits of EVA over an extended time period. This would suggest that valid examination of EVA should incorporate extensive time-series data with some allowance for these behavioural modifications. It is with these considerations in mind that the present study is undertaken. The paper itself is divided into four main areas. Section 2 outlines the concept and calculation of EVA and other accounting-based measures of firm value. The advantages and disadvantages of EVA are also briefly discussed. Section 3 deals with the empirical methodology employed in the paper, and the results are dealt with in Section 4. The paper ends with some brief concluding remarks in Section 5.

5 5 2. The Concept of Economic Value Added Despite the relatively recent adoption of EVA as an internal and external financial performance measure, its conceptual underpinnings derive from a well-established microeconomic literature regarding the link between firm earnings and wealth creation (Bell, 1998). For much of this history, at least since Alfred Marshall s Principles of Economics, the focus of analysis has been on adjustments to accounting earnings to reflect the opportunity cost of capital, largely since the unadjusted measure can be a misleading indicator of performance in both theory and practice. In the seminal contribution, Marshall (1990) states that the gross earnings of management which a man is getting can only be found after making up a careful account of the true profits of his business, and deducting interest on his capital. Later, the desirability of quantifying economic profit as a measure of wealth creation was operationalised by David Solomon (1965) as the difference between two quantities, net earnings and the cost of capital. This measure of residual income is then defined in terms of after-tax operating profits less a charge for invested capital which reflects the firm s weighted average cost of capital. Close parallels are thereby found in the related concepts of abnormal earnings, excess earnings ; excess income, excess realisable profits and super profits (Biddle et al., 1997). The comprehension of EVA as a practical and highly flexible refinement of economists concept of residual income the value that is left over after a company s stockholders (and all other providers of capital) have been adequately compensated provides the basis for the following discussion (Stern, Stewart and Chew, 1995, p. 32). The calculation of EVA itself consists of two separate but related steps. The primary adjustment is where a capital charge is subtracted from net operating profit after-tax (NOPAT). The capital charge is derived from multiplying the firm s overall financing cost, as reflected in the weighted average cost of capital (WACC), by the amount of invested capital. Invested capital in turn is defined as total assets, net of non-interest bearing current liabilities. In this form, EVA is essentially the same as residual income, though the latter measure is normally expressed as net income less a charge for the cost of equity capital (with the cost of debt already included in the calculation of net income) (Young, 1999). The second and more controversial step consists of a series of adjustments to GAAP-based numbers. Consisting of some 120 to 150 possible adjustments, these changes are made on the basis

6 6 of both empirical and theoretical concerns. First, it is argued that adjustments to accounting numbers are required in order to achieve higher correlations between the short term measure (in this case EVA), and share prices, which in turn can lead to more congruent goals for divisional managers and shareholders as well as a more reliable indicator of corporate performance for security analysts and portfolio managers (Young, 1999, p. 8). Second, at its root is the argument that not only are accounting earnings an inappropriate proxy for value creation, but that managers who are evaluated and compensated on the basis of earnings may take actions that increase earnings but destroy value, or fail to take actions that may reduce earnings but create value (Young, 1999, p. 8). Young (1999, p. 8) summarises the process as follows: These adjustments aim to 1) produce an EVA figure that is closer to cash flows, and therefore less subject to the distortions of accrual accounting; 2) remove the arbitrary distinction between investments in tangible assets, which are capitalised, and intangible assets, which tend to written off as incurred; 3) prevent the amortisation, or write-off, of goodwill; 4) eliminate the use of successful efforts accounting; 5) bring off-balance sheet debt into the balance sheets; and 6) correct biases caused by accounting depreciation. Notwithstanding the large number of adjustments that are possible, companies adopting the Stern Stewart EVA Financial Management System (as distinct from the publicly available performance rankings) generally make no more than fifteen adjustments to published accounts, though Young (1999) observes that this figure has progressively fallen in recent years. Young (1999, p. 9) attributes this decline to the reticence of corporate executives to diverge from GAAPbased numbers and the realisation that most of the proposed adjustments have little or no qualitative impact on profits. In the first place, the issues relating to the costliness of deviations from GAAP have been well documented. For example, Zimmerman (1997) argues that such deviations significantly increases both litigation costs (with the implication that the official financial statements are not really true and fair ) and administrative costs (through accounting systems becoming more complicated, costly to administer, and difficult for operating managers to comprehend). In the second instance, this would be consistent with Stewart s (1991) recommendation that adjustments should be limited to those situations where the outcomes are significant, where the item of interest may be reasonable influenced by management, and where the required information is readily available. The remaining modifications to a company s conventional accounts may be meaningfully grouped as adjustments to research and development, deferred taxes, intangibles, depreciation,

7 7 provisions for warranties and bad debts, restructuring changes, and macroeconomic conditions (Rennie, 1997). First, the EVA system is based on capitalising research and development, advertising and promotion expenditure over the estimated useful life of the expenditure, rather than expensing it as incurred. The argument follows that in a company which does not capitalise R&D, for example, managers might be tempted to under-invest because short-term profits will be adversely affected by R&D expenditures while the benefits will not be realised until future periods (Young, 1999). Second, taxes are only charged to profits as they are paid, rather than when they arise from timing differences between taxable income and book income under GAAP. The most significant source of the latter is the accelerated treatment of depreciation for tax purposes as against book income, with the argument that timing differences will recognise more book income than tax income (a deferred tax liability). Alternatively deferred tax assets arise when provisions are made for future costs that serve to reduce current book income. These may include provisions for warranties, restructuring and environmental clean-up. The net change of EVA is to add (or subtract) these changes in deferred tax to more accurately reflect the actual cash flows to tax authorities. Third, intangibles (especially goodwill arising from corporate acquisitions) are not automatically written-off in the EVA system. The argument is that the write-off of goodwill (whether at acquisition or more gradually through capitalisation and amortisation) effectively removes at least part of the acquirer s investment in the target from the balance sheet, thereby lessening management s burden to earn a competitive return on that portion of invested capital (Young, 1999, p. 12). Fourth, the key adjustment made to depreciation is that while a fast write-down for tax purposes is obviously preferred, the purported reality is that the operational value of plant does not fall as quickly as depreciation schedules suggest and management performance should reflect this [Joel Stewart as quoted in Rennie (1997)]. Stern Stewart also argues that profit-based measures provide a disincentive for new plant and efficiency gains through accelerated depreciation. Fifth, restructuring changes that involve cash payments are considered capital under EVA. The argument is that such adjustments should only be made to generate returns in a later period, and therefore should provide an appropriate return. Finally, adjustments are also made for extractive industries (mining, forestry, oil and gas). It is posited that broad performance changes in these industries result from the overly heavy emphasis on exogenous commodity prices: conditions over which management has little or no discretionary control.

8 8 A number of points may be made regarding the concept of EVA in general and the implementation of the EVA Financial Management System in particular. First, the notion of EVA is generally found to be not only akin to residual income in concept, it is also empirically comparable. This is the least contentious aspect of this increasingly widespread performance measure. Second, the adjustments made to GAAP-numbers represent a significant departure from traditional accounting-based measures such as residual income, earnings, and return on assets, investment and equity. In part, this has been the result of fundamental and well-documented limitations found in the accounting approach, especially the perceived biases and distortions in accounting numbers based on generally accepted accounting principles (GAAP). Key areas for EVA adjustment include provisions for deferred taxation, depreciation, the capitalisation and amortisation of goodwill, and research and development expenditure. Notwithstanding the intuitive appeal of many of these adjustments, Young (1999, p. 19), amongst others, argues that of far greater relevance to the corporate user than accuracy is whether the distortions that arise from GAAP result in suboptimal behaviour, and whether adjusting the numbers will improve managerial behaviour. This critique lies at the heart of the emerging EVA debate, especially given that the overriding objective is to more closely align management behaviour with improvements in shareholder value. While the case for adjusting EVA is more compelling for the external user, several studies have identified that for most firms adjusted EVA offers few advantages over unadjusted EVA (residual income) [see, for example, Biddle et al. (1997), Chen and Dodd (1997) and Clinton and Chen (1998)], and carries with it the costs of increased complexity and other costs that arise with departures from GAAP (Young, 1999). 3. Empirical Methodology Having illustrated EVA intuitively, we may extend the analysis to two closely related empirical questions. The first question relates to the purported dominance of EVA over both residual income and the conventional accounting performance measures of earnings before extraordinary items and net cash flow from operations in explaining contemporaneous stock returns. The second empirical question concerns those components unique to EVA that help explain these contemporaneous stock returns beyond that explained by residual income, earnings before extraordinary items and net cash flow from operations. Assuming that equity markets are (semi-strong form) efficient, stock market

9 9 returns may be used to compare the information content (or value-relevance) of these competing accounting-based performance measures in a regression-based approach (Bowen et al., 1987; Jennings, 1990; Easton and Harris, 1991; Ali and Pope, 1995; Biddle, Seow and Siegel, 1995; Sloan, 1996). Both relative and incremental information content comparisons are made. In terms of specific studies, the approach selected in the current study is most consistent with that used by Biddle et al. (1997) and Bao and Bao (1998), amongst others. The first methodological requirement is to describe the linkages that exist between the competing measures of firm performance; namely, earnings before extraordinary items (EBEI), net cash flow from operations (NCF), residual income (RI) and economic vale-added (EVA). Starting with EBEI as the most basic indicator of firm value we have: EBEI = NCF + ACC (1) where EBEI is the sum of net cash flow from operations (NCF) and accruals (ACC). ACC is defined as total accruals relating to operating activities and is composed of depreciation, amortisation, changes in non-cash current assets, changes in current liabilities, and changes in the non-current portion of deferred taxes. Net operating profit after tax (NOPAT) is a closely related indicator of current and future firm performance and is calculated by adding after-tax interest expense (ATI) to EBEI (1): NOPAT = EBEI + ATI = NCF + ACC + ATI (2) As indicated, the most significant difference between EBEI (1) and NOPAT (2) is that the later separates operating activities from financing activities by including the after-tax effect of debt financing (interest expense). As a measure of operating profit, no allowance is therefore made in (2) for the financing activities (both debt and equity) of the firm. One measure that does is residual income (RI) where operating performance is reduced by a net charge for the cost of all debt and equity capital employed: RI ( WACC CAP) = NCF + ACC + ATI CC = NOPAT (3) where WACC is an estimate of the firm s weighted average cost of capital, and capital (CAP) is defined as assets (net of depreciation) invested in going-concern operating activities, or equivalently, contributed and retained debt and equity capital, at the beginning of the period. The product of the firm s WACC and the amount of contributed capital thereby forms a capital charge

10 10 (CC) against which NOPAT is reduced to reflect the return required by the providers of debt and equity capital. A positive (negative) RI indicates profits in surplus (deficit) of that required by the suppliers of debt and equity capital and is associated with an increase (decrease) in shareholder wealth. The primary point of departure for EVA from RI (3) is the adjusting of both NOPAT and CAP for purported distortions in the accounting model of performance. EVA-type adjustments are made to both accounting measures of operating profits (NOPAT), and accounting measures of capital (CAP). EVA thereby reflects adjustments to GAAP in terms of both operating and financing activities. Simplifying, EVA is thus determined by: EVA = NCF + ACC + ATI CC + ADJ (5) where the total EVA accounting adjustment (ADJ) is the net figure of adjustments to NOPAT (NCF + ACC + ATI) less the adjustment to capital in determining CC (WACC CAP). Table 1 Selected descriptive statistics for firm valuation and EVA components pooled data ( ) Variable Mean Std. dev. Skewness Kurtosis MAR EVA EBEI NCF RI CC ATI ACC ADJ a MAR = compound annual stock returns lagged 3 months to fiscal year end; EVA = Stern Stewart measure of economic value-added; EBEI = earnings before extraordinary items; NCF = net cash flow; RI = residual income. b CC = cost of capital; ATI = after-tax interest; ACC = accruals; ADJ = Stern Stewart accounting adjustments; NCF = net cash flow. c All variables are expressed per share. The second methodological requirement is to specify the data sources for the financial statement numbers in equations (1) to (5) along with stock market returns as the dependent variable. Three separate sources of data are used. Data for EVA and its components are obtained directly from Stern Stewart s Australian EVA Performance Rankings. This data contains EVA, the weighted average cost of capital (WACC), return on capital (ROC), net operating profit after-tax (NOPAT),

11 11 capital (CAP) and average shareholder returns for Australia s 110 largest listed (non-financial) companies. The sample of firms consists of both adopters and non-adopters of the EVA Financial Management System over the period of Financial statement data for EBEI, ATI, RI, NCF, ACC and ADJ are collected from the Australian Stock Exchange s (ASX) Datadisk database and the Connect-4 Annual Report Collection database. Finally, share price data is obtained from the Australian Graduate School of Management s (AGSM) Share Price and Price Relative database (incorporating capitalisation adjustments and dividends). The final methodological requirement is to specify the models used to: (i) calculate the relative and incremental content of the competing measures of firm performance, and (ii) calculate the relative and incremental content of the components of economic value-added (EVA ) itself. The first specification is referred to as the firm valuation model : MAR b b EVA b EBEI b NCF b RI + e it = it + 2 it + 3 it + 4 it it (6) This model is estimated using a pooled time-series, cross-sectional least squares regression assuming cross-sectional heteroskedasticity and timewise autoregression (Greene, 1993). The dependent variable in (6) is the compounded annual stock market return (MAR). A 12-month nonoverlapping period ending three months following the firm s fiscal year end is chosen to allow time for information contained in the annual report to be impounded in market prices. The explanatory variables in the firm valuation model are net cash flows from operations (NCF), earnings before extraordinary items (EBEI), residual income (RI) and economic value-added (EVA). Both the dependent and explanatory variables are normalised by the number of outstanding shares. Selected descriptive statistics for these variables are given in Table 1 and Pearson and Spearman correlation coefficients are calculated in Table 2. Following the valuerelevance literature on financial statement information, this hypothesis suggests positive coefficients for EBEI, EVA, NCF and RI when used as an explanatory variable for market returns. The second specification examined is referred to as the components of EVA model: MAR b b CC b ATI b ACC b ADJ b NCF + e it = it + 2 it + 3 it + 4 it + 5 it it (7) This model is also estimated using a pooled time-series, cross-sectional least squares regression with corrections for heteroskedasticity and autocorrelation. The dependent variable is given as MAR. The independent variables are the five components of EVA: namely, net cash flows (NCF), operating accruals (ACC), after-tax interest (ATI), cost of capital (CC) and accounting

12 12 adjustments (ADJ). All variables are as previously defined. Descriptive statistics are provided in Table 1 and correlation coefficients in Table 2. Both EVA and the set of explanatory variables are normalised by the number of shares on issue. The first variable ACC is defined as earnings less net cash flow from operations (EBEI - NCF). Accruals can either be positive or negative, but are usually negative (reflecting non-cash expenses such as depreciation and amortisation). The ex ante sign on the coefficient for accruals is thought to be positive when specified as an explanatory variable for market returns. ATI is calculated as one minus the firm s tax rate (assumed to be 36 per cent) multiplied by interest expense. A positive coefficient is hypothesised when market returns are regressed against interest expense. Table 2 Pearson (product moment) and Spearman (rank) correlation coefficients for firm valuation and EVA components pooled data ( ) Firm valuation variables MAR EVA EBEI NCF RI MAR EVA EBEI NCF RI EVA components MAR CC ATI ACC ADJ NCF MAR CC ATI ACC ADJ NCF a MAR = compound annual stock returns lagged 3 months to fiscal year end; EVA = Stern Stewart measure of economic value-added; EBEI = earnings before extraordinary items; NCF = net cash flow; RI = residual income. b CC = cost of capital; ATI = after-tax interest; ACC = accruals; ADJ = Stern Stewart accounting adjustments; NCF = net cash flow. c Cells above the diagonal contain Spearman rank correlation coefficients and those below the diagonal contain Pearson correlation coefficients. CC is defined as each firm s weighted-average cost of capital times its beginning of year capital (WACC CAP). A negative coefficient is hypothesised. Finally ADJ reflects Stern Stewart s adjustments to earnings and capital, and is defined as economic value-added less residual income (EVA - RI). Given the fact that the direction of change for ADJ may vary across firms in the sample depending on both financing and operations (GAAP-related accounting

13 13 adjustments can either be positive or negative), it is somewhat difficult to postulate the relationship between GAAP adjustments and market returns. No a priori coefficient is postulated. As discussed, tests of these models are based on incremental and relative information specification tests. The first sets are joint hypothesis tests that EVA, RI, NCF and EBEI have equal relative information content. The comparisons of the R 2 of the regression results are made to determine which variable better explains variation in MAR. Rejection of this hypothesis is viewed as evidence of a significant difference in the relative information content. The second set of tests indicates whether one of these predictors of firm value provides value-relevance data beyond that provided by another. Rejection of this hypothesis is viewed as evidence of incremental information content. Similar tests of relative and incremental information content are performed in the components of EVA model. The components in this instance are EVA, NCF, ACC, ATI, CC and ADJ. 4. Empirical Results Table 3 provides the estimated coefficients and standard errors of the firm valuation model discussed in the previous section. The dependent variable is specified as compounded annual market returns (with a lagged period of three months following fiscal year end) and the explanatory variables are variously specified as economic value-added, earnings before extraordinary items, net cash flow and residual income. An assumption of a linear relationship between these variables is made. Both the firm valuation model (Tables 3 and 4) and the EVA components model (Tables 5 and 6) are tested for multicollinearity (autocorrelation and heteroskedasticty having being corrected in the estimation technique). The auxiliary R 2 of per cent suggests that multicollinearity while present, is not excessively high, especially considering the nature of the microeconomic data. The EVA components model obtains a high R 2 when testing for multicollinearity which suggests that this would be (obviously) a problem were the emphasis on individual coefficients rather than explanatory power. The significance of the estimated coefficients in Table 3 suggests that all four accounting-based performance measures are positively associated with market returns over the period Of the twenty estimated slope coefficients, seventeen are significant at the.10 level or lower and only two are not in the predicted direction (both NCF). This finding holds even when a pairwise

14 14 combination of performance measures are specified in the same regression. The summary results of these regressions in the form of relative and incremental information content tests are presented in Table 4. Table 3. Association with market-adjusted returns for the firm valuation model, pooled time-series cross-sectional data CONS. EVA EBEI NCF RI F R 2 *** *** (0.0501) (0.0223) *** *** (0.0219) (0.0478) *** *** (0.0233) (0.0433) *** *** (0.0251) (0.0336) *** *** *** (0.0195) (0.0190) (0.0508) *** *** *** (0.0201) (0.0201) (0.0447) *** *** (0.0258) (0.0306) (0.0492) *** *** (0.0218) (0.0711) (0.0564) *** *** *** (0.0241) (0.0452) (0.0183) *** *** *** (0.0253) (0.0400) (0.0292) *** *** * (0.0235) (0.0301) ( (0.0549) (0.0375) a Asterisks indicate level of significance at the *** 0.01, ** 0.05 and * 0.10 level. Figures in brackets are standard errors. b Estimated coefficients are from Eq. (6): MAR it = b 0 + b 1 EVA it + b 2 EBEI it + b 3 NCF it + b 4 RI it + e it, where MAR it = compound annual stock returns lagged 3 months to fiscal year end; EVA it = Stern Stewart measure of economic valueadded; EBEI it = earnings before extraordinary items; NCF it = net cash flow; RI it = residual income. Part A of Table 4 indicates that there is a significant difference in relative information content. The highest adjusted R 2 from the single coefficient regressions is shown on the left, with lower explanatory power in descending order to the right. The suggestion is that EBEI better explains MAR than does EVA, NCF and RI alone. In terms of international comparisons, Biddle s et al. (1997) results also indicated that earnings (EBEI) was more highly associated with market-adjusted returns than either RI or EVA, but that all three measures dominate net cash flow (NCF). In this

15 15 study, the relative information content of EVA is the lowest of the four accounting-based performance measures, accounting for only some percent of variation in market returns. Put differently, when stock returns are specified as the dependent variable, EVA accounts for only some 60.3 percent of the variation that EBEI does, 74.1 percent as found in RI, and 78.9 percent of the variation attributed to NCF. Notwithstanding the low explanatory of accounting-based measures in general and EVA in particular, the estimated coefficients are highly significant, with joint hypothesis F-tests that all slope coefficients are zero rejected at the.05 level or better. Furthermore, the explanatory power of all four accounting-based measures is significantly higher than that found in a number of comparable studies. For example, Biddle et al. (1997) estimated the relative information content of EBEI, RI, EVA and NCF at 9.04, 6.24, 5.07 and 2.38 percent respectively. Table 4 Relative and incremental information content for firm valuation model A. Relative Information Content EBEI > RI > NCF > EVA 23.67% 19.29% 18.10% 14.29% B. Incremental Information Content EVA/ EBEI/ EVA/ NCF/ EVA/RI RI/EVA EBEI/ NCF/ EBEI/RI RI/EBEI NCF/RI RI/NCF EBEI EVA NCF EVA NCF EBEI 0.88% 10.26% 2.26% 6.07% 0.07% 5.07% 5.76% 0.19% 4.95% 0.57% 1.98% 3.17% a The incremental information content for EVA/EBEI, for example, is calculated as the R 2 from the pairwise regression minus the individual R 2 for EBEI. b Underlying regressions are from Eq. (6): MAR it = b 0 + b 1 EVA it + b 2 EBEI it + b 3 NCF it + b 4 RI it + e it, where MAR it = compound annual stock returns lagged 3 months to fiscal year end; EVA it = Stern Stewart measure of economic valueadded; EBEI it = earnings before extraordinary items; NCF it = net cash flow; RI it = residual income. The results in Part B of Table 4 are also based on (6) and provide incremental information content tests for the pairwise combinations of EVA, EBEI, RI and NCF. For example, EVA/EBEI (0.88 percent) is equal to the information content of the pairwise comparison of EVA and EBEI (24.55 percent) minus the information content of EBEI (23.67 per cent). The pairwise combinations of EVA and EBEI, NCF and RI indicates that explanatory power has increased by 10.26, 6.07 and 5.07 respectively over the EVA measure alone. A comparison with the incremental information tests contained in Bao and Bao (1998) for pooled data indicates that earnings have a zero impact on EVA alone, while residual income increases explanatory power by some 38 percent. Overall, the results indicate that EBEI exhibits the largest relative information content among the measures, with EVA (0.88 percent), RI (0.57 percent) and NCF (0.19 percent) providing only limited

16 16 incremental information content beyond earnings. The most logical pairing of information variables in explaining market returns is therefore composed of EBEI and EVA. Table 5. Association of market returns with components of economic value-added, pooled time-series cross-sectional data ( ) CONS. CC ATI ACC ADJ NCF F R 2 *** *** (0.0223) *** *** (0.0225) (0.0392) *** *** (0.0231) (0.0482) *** *** (0.0650) (0.0317) *** *** (0.0641) (0.0458) *** *** *** (0.0213) (0.0246) (0.0388) *** *** *** (0.0218) (0.0274) (0.0532) *** *** (0.0252) (0.0495) (0.0622) *** *** *** (0.0218) (0.0265) (0.0449) *** ** *** (0.0228) (0.0465) (0.0634) *** *** *** (0.0242) (0.0296) (0.0221) *** *** *** (0.0619) (0.0448) (0.0508) *** *** (0.0652) (0.0547) (0.0231) *** *** (0.0155) (0.0644) (0.0546) *** ** *** (0.0665) (0.0239) (0.0459) *** * ** ** (0.0249) (0.0492) (0.0371) (0.0625) (0.0559) (0.0253) a Asterisks indicate level of significance at the *** 0.01, ** 0.05 and * 0.10 level. Figures in brackets are the standard errors. b Estimated coefficients are from Eq. (7): MAR it = b 0 + b 1 CC it + b 2 ATI it + b 3 ACC it + b 4 ADJ it + b 4 NCF it + e it, where MAR it = compound annual stock returns lagged 3 months to fiscal year end; CC it = capital charge; ATI it = after-tax interest; ACC it = accruals; ADJ it = Stern Stewart accounting adjustments; NCF it = net cash flow.

17 17 The second phase of the study is to examine the components of EVA. These components are the cost of capital (CC), after-tax interest (ATI), accruals (ACC), accounting adjustments (ADJ) and net cash flows from operations (NCF). Table 5 presents the results of the individual and pairwise regressions of the components on EVA. In these regressions the estimated coefficients for CC, ATI, ACC, ADJ and NCF are all significant at the 0.01 level in the individual variable regressions. The measure for capital charge adjustments (CC) is insignificant when paired with operating adjustments (ADJ) suggesting a high degree of collinearity between the two steps of EVA GAAP adjustments. In the final regression specification in Table 5 the ex post signs for CC (negative) and NCF (positive) do not correspond with a priori reasoning, while the figure for Stern Stewart operating performance adjustments (ADJ) is insignificant. In none of the single coefficient or pairwise regressions does the estimated sign for ATI and CC correspond with the ex ante sign (both negative). Notwithstanding this result, all of the one-tail F-tests are significant at the.05 level or better, thereby rejecting the null hypothesis of the joint insignificance of the slope coefficients. Table 6 Relative and incremental information content for the EVA model A. Relative Information Content ACC > ATI > CC > NCF > ADJ 20.50% 19.47% 17.81% 13.12% 8.13% B. Incremental Information Content CC/ATI ATI/CC CC/ACC ACC/CC CC/ADJ ADJ/CC CC/NCF NCF/CC ATI/ACC ACC/ATI 2.94% 4.60% 2.40% 5.09% 11.67% 1.99% 8.49% 3.80% -0.89% 1.92% ATI/ADJ ADJ/ATI ATI/NCF NCF/ATI ACC/ADJ ADJ/ACC ACC/NCF NCF/ACC ADJ/NCF NCF/ADJ 13.69% 2.35% 1.84% -4.51% 7.66% -4.71% 2.67% -4.71% 0.39% 5.38% a Underlying regressions are from Eq. (7): MAR it = b 0 + b 1 CC it + b 2 ATI it + b 3 ACC it + b 4 ADJ it + b 5 NCF it + e it, b The incremental information content for ATI/ADJ for example is calculated as the R 2 from the pairwise regression minus the individual R 2 for ATI. Part A of Table 6 gives the results of relative information content tests of the components of EVA. When specified as a single slope coefficient ACC (20.50 percent) has greater explanatory power than ATI (19.47 percent), CC (17.81 percent), NCF (13.12 percent) and ADJ (8.13 percent). This is consistent with the previous part of the analysis since ACC is shared by EVA with EBEI, RI and NCF, ATI and CC with RI, and ADJ by itself alone. Part B of Table 6 presents the incremental information content results. Starting with the base NCF, ACC adds 2.67 percent in explanatory power, ATI adds 1.84 percent, CC 8.49 percent and ADJ only 0.39 percent. Overall, the component

18 18 of EVA that explains most variation in stock returns is accruals, followed by after-tax interest, capital charges, net cash flow and accounting adjustments. This also highlights the results obtained in the first part of the analysis, where most variation in returns is explained by conventional accounting-based measures of performance, though EVA vis-á-vis GAAP-related accounting adjustments are still significant at the margin. 5. Concluding Remarks A number of points emerge from the present study. The first part of the analysis uses pooled time-series, cross-sectional data of 110 listed Australian companies to evaluate the usefulness of EVA and other accounting-based performance measures. The measures of relative and incremental information content indicate that over the period 1992 to 1998 no more than 25 percent of the variation in stock returns could be explained by these measures. Of that explanatory power, the bulk was encompassed in conventional earnings, with relatively minor improvements in explanatory power associated with the inclusion of accruals, after-tax interest and capital charges and GAAP-based accounting adjustments. Notwithstanding the obvious importance of earnings figures in value-relevance studies, EVA is still significant at the margin in explaining variation in market returns. This would support the potential usefulness of EVAtype measures for internal and external performance measurement. In the second part of the paper, the components of EVA are specified as explanatory variables in regressions with market returns. When examining the components of EVA (most of which are shared with related performance measures), after-tax interest (ATI) and accruals (ACC) are found to be the most significant components. This is followed by the capital adjustments (CC) associated with EVA, and finally the operating activity GAAP-related adjustments (ADJ). All other things being equal, variation in stock returns is largely explained by commonly available financial statement variables, though the adjustments most closely associated with EVA, namely GAAP-related accounting adjustments and capital charges, are still significant at the margin. There are at least three ways in which this research may be extended. First, a limitation in this study is that a comparison could not be made of firms who use the EVA Financial Management System (incorporating redesigned executive compensation plans) against firms that use traditional accounting earnings-based incentives. While the results in the present study are suggestive of the

19 19 benefits of EVA as a tool for internal performance measurement and compensation design, it is conceivable that the association between EVA and returns is even stronger for EVA adopters (Biddle et al., 1997; Wallace, 1997; Ferguson and Leistikow, 1998). Second, there is abundant empirical evidence to suggest that models relating accounting and market returns have more explanatory power when the accounting returns are expressed by relative changes and the relation is a non-linear, convex-concave function [see, for example, Freeman and Tse (1992), Riahi- Belkaoui (1996) and Frankel and Lee (1998)]. A further avenue of research would therefore consist of alternative specifications of accounting and market returns, along with the use of nonlinear regression techniques. Finally, there is scope for the investigation of the usefulness of EVA as an internal and external performance measure in other settings. Stern Stewart also provide performance rankings for listed companies in the U.K., Canada, Brazil, Germany, Mexico, Turkey, and France, and empirical evidence from these institutional milieus would provide additional evidence regarding the contextual and/or substantive usefulness of accountingbased value-added measures. References Ali, A. and Pope, P.F., The incremental information content of earnings, funds flow and cash flow: The UK evidence. Journal of Business Finance and Accounting 22(1), Bao, B-H., Bao, D-H., Usefulness of value added and abnormal economic earnings: An empirical examination. Journal of Business Finance and Accounting 25(1-2), Bell, L.W., Economic profit: An old concept gains new significance. Journal of Business Strategy 19(5), Biddle, G.C, Bowen, R.M., and Wallace, J.S., Does EVA beat earnings? Evidence on associations with stock returns and firm values. Journal of Accounting and Economics 24(3), Biddle, G.C., Seow, G.S., and Siegel, A.F., Relative versus incremental information content. Contemporary Accounting Research 12(1), Bowen, R.M., Burgstahler, D., and Daley, L.A., The incremental information content of accrual versus cash flows. The Accounting Review 62(4), Chen, S. and Dodd, J.L., Economic value added (EVA): An empirical examination of a new corporate performance measure. Journal of Managerial Issues 9(3), Clinton, B.D., and Chen, S Do new performance measures measure up? Management Accounting 80(4), Easton, P.D. and Harris, T.S., Earnings as an explanatory variable for returns. Journal of Accounting Research 29(1), Ferguson, R., and Leistikow, D., Search for the best financial performance measure: Basic are better. Financial Analysts Journal 54(1), Frankel, R. and Lee, C.M.C Accounting valuation, market expectation, and cross-sectional stock returns. Journal of Accounting and Economics 25, Freeman, R.N. and Tse, S.Y A nonlinear model of security price responses to unexpected earnings. Journal of Accounting Research 30(2), Greene, W.H., Econometric Analysis. Prentice-Hall, New Jersey.

20 20 Herzberg, M.M., Implementing EBO/EVA analysis in stock selection. Journal of Investing, Spring, Jennings, R., A note on interpreting incremental information content. The Accounting Review 65(4), Lee, C., Measuring wealth. The CA Magazine, April, Lehn, K., and Makhija, A.K., EVA, accounting profits, and CEO turnover: An empirical examination, Journal of Applied Corporate Finance 10(2), Marshall, A., Principles of Economics, Porcupine Press, 8 th Edition (1920), Macmillan: London, 1990 Reprint. O Byrne, S.F., EVA and market value. Journal of Applied Corporate Finance 9(1), Rennie, P., Finding the best, and worst, wealth creators. Business Review Weekly 19(30), Riahi-Belaoui, A., Earnings-returns relation versus net-value-added-returns relation: The case for a nonlinear specification. Advances in Quantitative Analysis of Finance and Accounting 4, Rogerson, W.P., Intertemporal cost allocation and managerial investment incentives: A theory explaining the use of economic value added as a performance measure. Journal of Political Economy 105(4), Ross, I., The Stern Stewart performance Journal of Applied Corporate Finance 11(4), Sloan, R.G., Do stock prices fully reflect information in accruals and cash flows about future earnings? The Accounting Review, Solomons, D., Divisional Performance: Measurement and Control, Homewood, IL: Irwin. Stern Stewart and Co., WWW < Accessed February Stern, J.M., Stewart, G.B. and Chew, D.H The EVA financial management system. Journal of Applied Corporate Finance 8(2), Stewart, G.B., The Quest for Value. New York: Harper. Stewart, G.B., EVA: Fact and fantasy. Journal of Applied Corporate Finance 7(2), Uyemura, D.G., Kantor, C.C. and Petit, J.M EVA for banks: Value creation, risk management, and profitability measurement. Journal of Applied Corporate Finance 9(2), Wallace, J.S., Adopting residual income-based compensation plans: Do you get what you paid for? Journal of Accounting and Economics 24, Wood, C., Value-based management tools gain some leverage. Business Review Weekly 18(49), Young, S.D., Some reflections on accounting adjustments and economic value added. Journal of Financial Statement Analysis 4(2), Zimmerman, J.L., EVA and divisional performance measurement: Capturing synergies and other issues. Journal of Applied Corporate Finance 10(2),