WestminsterResearch

http://www.westminster.ac.uk/westminsterresearch

EVA and shareholder value creation: an empirical study Wajeeh Elali

Westminster Business School

This is an electronic version of a PhD thesis awarded by the University of Westminster. The Author, 2007. This is a scanned reproduction of the paper copy held by the University of Westminster library.

The WestminsterResearch online digital archive at the University of Westminster aims to make the research output of the University available to a wider audience. Copyright and Moral Rights remain with the authors and/or copyright owners. Users are permitted to download and/or print one copy for non-commercial private study or research. Further distribution and any use of material from within this archive for profit-making enterprises or for commercial gain is strictly forbidden.

Whilst further distribution of specific materials from within this archive is forbidden, you may freely distribute the URL of WestminsterResearch: (http://westminsterresearch.wmin.ac.uk/). In case of abuse or copyright appearing without permission e-mail repository@westminster.ac.uk

EVA and Shareholder Value Creation: An Empirical Study

Wajeeh Elali

A thesis submitted in partial fulfilment of the of the University of Westminster requirements for the degreeof Doctor of Philosophy

September2007

DEDICATION

This thesisis dedicatedto my belovedfather

who supported, inspired, and encouraged me to pursue my education to my

fullest abilities.
He was always my greatfriend and mentor!

ii

ACKNOWLEDGEMENTS

First and foremost, I must acknowledge and thank my supervisory team to whom deeply indebted. I am particularly grateful to Dr. Kadom Shubber and Dr. Nic am Zafiris, not only for their guidance and expertise, but also for their assistance, facilitated my research work for this understanding, and accessibility, which greatly thesis. Their patience and trust in my being able to accomplish this task, despite my often daunting work load, were an encouraging and value added factor, to say the least. I truly benefited from them individually and enjoyed working with them over the last five years. I have been truly fortunate to have them as my supervisors as they are the kind of people that one would like to cultivate as friends as well as mentors. I wish also to extend my thanks and sincere appreciation to the Board of Examiners: Professor David Weir, Professor Ben Nowman, and Dr. Peter Urwin (Chairman). Their constructive critiques and insightful recommendationswere invaluable in completing this project. I would further like to acknowledge and express my thanks to Mike Fisher (The University Registrar), Patricia King-Edwards, Cristian Popescu, Eric Chang for their cooperation and professional assistance.Finally, I owe a special acknowledgement and thanks to my sister Dr. Hanan Ismail, my brother in-law Dr. Khalil Muhdi, and my niece Noor for making their home available for my stays in London. Their encouragementand hospitality gave me peace of mind during the various stagesof my thesis.

ABSTRACT
In recent years, a variant of residual income often called Economic Value Added (EVA)' or Economic Income (EI) has become a popular concern in academia and business communities. This study investigates the general hypothesis that EVA is more highly associated with firm values than are traditional shareholder wealth and

measures. Two commonly used value-based performance metrics namely, performance Total ShareholderReturn (TSR) and Tobin's Q are also consideredto highlight the valueEVA vis-a-vis these measuresin predicting shareholderwealth. relevanceof Using a sample of panel data of around 12,000 firm-year observations taken from the Stem Stewart 1000 EVA/MVA databaseand the DATASTREAM file over the period 1991-2002, this study finds compelling evidence that shareholder value is a function of EVA. This study also provides evidence consistent with the notion that EVA outperforms traditional performance measures in explaining shareholder wealth. Valueother tests reveal EVA to be more highly associatedwith shareholder wealth than relevance TSR and Tobin's Q. The incremental tests also suggest that EVA possesses the largest (or information usefulness) over TSR and Tobin's Q. These results explanatory power the claims made by EVA proponents and further support the conclusively support the EVA metric for internal and external performance. potential usefulnessof

' EconomicValue Added or EVA is a relatively new measureof corporateperformancedevelopedand Stern Stewartand Co. (hereafter in trademarked the late 1980sby the US-basedbusiness consultants to as Stem Stewart). referred

iv

TABLE OF CONTENTS
DEDICATION ACKNOWLEDGEMENTS ABSTRACT LIST OF FIGURES LIST OF TABLES CHAPTER 1: INTRODUCTION 1.1 Objective of the Study 1.2 ResearchQuestions and Hypotheses 1.3 Outline of Thesis 1.4 Concluding Remarks CHAPTER 2: BACKGROUND AND LITERATURE REVIEW 2.1 Introduction 2.2 ShareholderValue Approach 2.3 Value-based Metrics 2.3.1 Economic Value Added (EVA) 2.3.2 Market Value Added (MVA) 2.3.3 Tobin's Q 2.3.4 Total ShareholderReturn JSR) 2.3.5 Cash Flow Return on Investment (CFROI) 2.3.6 The Balanced Scorecard(BSC) 2.4 The Association between EVA & ShareholderValue 2.5 Concluding Remarks CHAPTER 3: RESEARCH ISSUES and HYPOTHESES 3.1 Introduction 3.2 ResearchIssues 3.3 ResearchHypotheses 3.3.1 Hypothesis One 63 64 65 65 7 7 19 20 35 39 43 45 50 53 61 iv vii viii I 3 4 5 5

3.3.2 Hypothesis Two 3.3.3 Hypothesis Three 3.4 Concluding Remarks

66 67 68

CHAPTER 4: RESEARCH DESIGN & METHODOLOGY 4.1 Introduction 4.2 Data Sources and Statistical Techniques 4.3 Model Specification 4.4 ResearchVariables 4.4.1 Dependent Variable 4.4.2 Independent Variables 4.5 Concluding Remarks CHAPTER 5: DATA ANALYSIS AND PRESENTATION OF RESULTS 5.1 Introduction 5.2 Descriptive Statistics 5.3 Test of Hypothesis 5.3.1 Hypothesis One 5.3.2 Hypothesis Two 5.3.3 Hypothesis Three 5.4 Concluding Remarks 80 81 82 82 85 91 99 69 69 72 74 74 75 79

CHAPTER 6: CONCLUSIONS, LIMITATIONS, AND FUTURE RESEARCH 101

REFFERENCES APPENDICES Appendix A: Figures Appendix B: Tables Appendix C: The 2002 Stem Stewart Performance 1000

107 116 117 121 165

vi

LIST OF FIGURES

Figure One: Figure Two:

The Shareholder Value Analysis Network A Typical Financial Management System (Fuzzy Finance)

117 118 119 120

Figure Three: EVA: A Simplified and FocusedFinancial Management System Figure Four: Balanced Scorecard(Casual Chain)

vii

LIST OF TABLES

Table 2.1 Table 4.1 Table 4.2 Table 5.1

How Performance Measures"Explain" Changesin MVA Calculation of NOPAT from Financial StatementData Calculation of Capital Using Accounting Financial Statements Descriptive Statistics of Variables Employed in the RegressionModels

121 122 123

124 125 126 128 130 133 136 139 140 143 148 153 163 164

Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 5.10 Table 5.11 Table 5.12 Table 5.13 Table 5.14

Correlation between Variables Hypothesis (1): Univariate RegressionResults Hypothesis (2): Multivariate RegressionResults Fixed vs. Random Effects (Equation 5.3) Statistical Results of Step One of Hypothesis (2) Testing Statistical Results of Step Two of Hypothesis (2) Testing Hypothesis (2): Relative & Incremental Value-relevance Tests Hypothesis (3): Multivariate RegressionResults Fixed vs. Random Effects (Equation 5.13) Statistical Results of Step One of Hypothesis (3) Testing Statistical Results of Step Two of Hypothesis (3) Testing Hypothesis (3): Relative Value-RelevanceTest Hypothesis (3): Incremental Value-RelevanceTest

viii

Chapter 1 INTRODUCTION
The link between performance measures and shareholder value creation has

interest.Academicresearchers, becomean issueof considerable academic practitioner and

business analysts have engaged in a rather heated debate in the corporate executives, and last decade or so as to whether the new value-based performance metrics have a higher their returns than do other traditional accounting-based correlation with stock values and 2 Economic Value Added (or EVA), the income remaining after all residual measures. costs, including the opportunity cost of the equity capital employed, is among the few that have been widely adopted and are claimed to approximate performance metrics In effect, EVA is promoted by its proponents as being superior to shareholder returns. traditional and non-traditional performance metrics as a determinant and predictor of other (Stewart, 1991; Ehrbar, 1998). corporate successand value creation Nevertheless, despite the growing amount of literature that has attempted to the claims made about EVA's superiority; little empirical researchhas so far been evaluate done to support the above assertions (e.g., Ittner & Larcker, 1998; Lehn and Makhija, 1997; Lovata and Costigan, 2002; Yook, 1999; Feltham et al., 2004). Moreover, the limited studies that have appeared in the literature have produced somewhat conflicting For instance, Biddle et al. (1997); Chen and Dodd (1997); Fernandez(2002); conclusions.

2 See, for example, Ittner and Larcker (1998 and 2001); Black et al. (2001); Arnold and Davies (2000); Garvey and Milbourn (2000); Myers (1996); Chen and Dodd (1997); Biddle et al. (1997); Worthington and West (2004); Erasmusand Lambrechts(2006); Rajan (2000); Feltham et al. (2004); Fergusonet al. (2005); Fabozzi and Grant (2000). and

Paulo (2003); Palliam (2006); and Stark and Thomas (1998) have mostly not been

supportive of these claims. On the other hand, Grant (1996,2003); Lehn and Makhija
(1996,1997); Zafiris and Bayldon (1999); Young and O'Byrne (2001); Worthington and West (2004); Tully (1993,1998); Ferguson et al. (2005); Erasmus and Lambrechts (2006); and Feltharn et al. (2004) have made contributions that favour EVA on theoretical and/or empirical grounds. The inconclusive and mixed results of these studies raise an important question. Is EVA really superior to other alternative performance measures or is it merely a fad by a management consultancy firm? This conflicting evidence thus necessitates promoted the conducting of further studies that may provide better insight and understanding into this complex, yet crucial relationship between shareholder wealth creation and EVA. To further this idea, Lovata and Costigan (2002, p.226) stated, "Economic Value Added is a that requires much additional research to support or contest the claims of its concept developers. Likewise, Feltham et al. (2004, p. 83) suggests that the debate should be " EVA has greater relevancethan other performance measures. reopenedregarding whether On the other hand, there has been an emphasis in previous empirical work in this limited set of panel data. For example, area on either a cross-section of companiesor on a Bao and Bao (1998) only employed a cross-sectionof 166 firms over the period of 199293; whereas, Grant (2003) focused on only 50 of the largest U. S. wealth

2000. Clearly, an examination of extended "panel data" creators/destroyers at year-end the usefulness of EVA as an would certainly permit greater empirical certainty on

advanced measure of corporate performance and value creation. Thus, the scope of this study must out of necessity go further. This study, therefore, explores the suitability of using EVA as a measure of corporate success as well as providing additional empirical evidence on the use of EVA. Specifically, the statistical association between EVA and the creation of shareholders has been empirically examined and highlighted. The efficiency of two alternative wealth value-basedperformance measures-namely, Total ShareholderReturn (TSR) and Tobin's was also considered vis-a-vis EVA to assertor refute its superiority. In this chapter, the overall objectives of the thesis are highlighted, and are followed by a brief discussion of the three research questions that form the foundation of this study and so comprise the basis for its hypotheses.

1.1 Objectives of the Study

The primary objective of this thesis is to empirically test the assertion that EVA is highly associatedwith Market Value Added (or MVA). Market Value Added is defined as the difference between the market value of the firm (including equity and debt) and the total capital invested in the firrn (Young and O'Byrne, 2001, p.29). It is a measure of is considered to be the best indicator of shareholder value external performance, which The study does not seek, though, to fully explain the determinants of MVA, but creation. to show how well EVA acts as a genuine explanatory variable for MVA, in order to only justify its usefulness for performance measurement,shareholder value creation, executive financial reporting. compensation,and

Thus, the objectives of this study are three-fold. First, to ascertain whether there is a significant statistical association between EVA and shareholder wealth. Second, to examine which value-based performance metric (EVA, TSR, or Tobin's Q) has a greater association with market value added (MVA). Finally, to attempt to uncover whether the components unique to EVA-namely, net operating profit after tax (NOPAT), return on

invested capital (ROIC), profitability index (PI), Capital Growth (CG), cost of capital (WACC), and total invested capital (TIC) -- help in explaining contemporaneousMVA beyond that explained by traditional value-basedperformance measures.

1.2 Research Questions and Hypotheses

The hypothesesto be tested are derived from the notion that EVA is generally more highly associated with shareholder returns and firm value than are other alternative performance measures.Thus, this study posesthe following questions: e Does a statistical relationship between EVA and shareholderwealth in the sense of contemporaneousMarket Value Added (MVA) exist, and if it does, how much of the variation (i. e., change) of the shareholder value be explained by EVA? can * Does EVA dominate other commonly-used value-based measures in explaining MVA? 9 Do components unique to EVA help to explain contemporaneousMVA beyond that explained by traditional value-basedperformance measures?

These three questions form the foundation of this study and consequently comprise the basis for its hypotheses.While the first question explores the direction and the strength of the relationship between EVA and MVA; the second and third one's investigate the EVA efficiency of vis-a-vis other alternative performance metrics in predicting

shareholder value.

1.3 Outline of Thesis

To addressthese researchquestions,this study is organized as follows: In Chapter 2, the relevant literature on value-basedperformance metrics and shareholder is reviewed and discussed. Chapter 3 presents the empirical research value creation hypotheses.Chapter 4 describes the population and data sourcesas well as presenting the design and methodologies for testing the hypotheses.In Chapter 5, the results of a research investigation are presentedfor each hypothesis and discussedin comprehensive statistical detail. Finally, chapter 6 summarizesthe statistical results and discussesadditional aspects this study for future research. of

1.4 Concluding Remarks

Over the last two decadesor so, there has been a growing concern among business and professional managers that traditional accounting measuresof analysts, academics, longer appropriate for the purpose of strategic decisions and control. In performance are no the ongoing search for more adequate performance measures that show some link to

shareholder value, a number of alternativevalue-based metrics have been developedand

by their respective advocates. The EVA, which is a residual income metric that promoted subtracts the cost of capital from the operating profits generatedin the business,seemsto have emerged as a real improvement over the traditional accounting measures.This study the assertion that EVA is a superior measureof performance and value creation. explores In the following chapters, this issue will be thoroughly investigated and empirically

examined.

Chapter 2 BACKGROUND
2.1 Introduction
The association between alternative performance measures and firm value has deal of academic interest for a long period of time. Several explanations attracted a great and predictions regarding this relationship have been raised in both the economic and literature. The central topic addressedin this chapter involves how the value accounting creation process of the firm. is reflected in alternative value-based performance metrics. The purpose of this chapter is to provide the basic theoretical and empirical foundation for the thesis as well as give a detailed review of six of the most widely used and discussed in the business world and academic literature that is, value-basedperformance metrics Economic Value Added (EVA); Market Value Added (MVA); Cash Flow Return on Investment (CFROI); Market-to-Book Value Ratio (MBV) better known as Tobin's Total Shareholder Return (TSR); and the Balanced Scorecard (BSC). The specific between these performance metrics and shareholder value creation will also relationships be discussedand highlighted. Concluding remarks are provided at the end of the chapter.

AND LITERATURE

REVIEW

2.2 Shareholder Value Approach

Over the last decade or so and in particular globalization, de-regulation, trade liberalization, technological changes,transparencyor fuller disclosure of activities, and the information revolution including the internet, as well as the increasing sophistication of the

financial markets have become the dominant forces behind the transformation of
corporations and the climate in which they operate. Companies across the globe are now under unprecedentedpressureto adapt to this new climate and to perform consistently well in all markets, in which they compete the product market, the labour market, and -namely, the capital market. Otherwise, they would be out of business. What will corporate successlook like in the decadesahead?Certainly, the rules of the game have changed. Corporations are now finding that making good products and trying desperately to satisfy customers are still necessary,but are no longer sufficient. Nor is it enough to focus on traditional earning figures alone. The key to successin today's business environment is the simultaneous delivery of a superior return to investors, focusing on core businesses,removing constraints, doing proactively managing risks, businessdifferently, as well as maintaining sustainablegrowth rates. A team of experts at Price Waterhouse Coopers aptly declared that, even though, these challenges may look new; essentially, they are the same old ones in a different guise (Black et al., 2001, p.24). The three aspects of market activity, that is risk, growth, and to be crucial. The fact that cash is king and investors require an adequate return, continue for the risks they bear, has not changed. What has changed, though, is the compensation focus. Despite the fact that the capitalist system has always seemed to be structured to the interest of the owners of the equity (i. e., the shareholders); it is only now, with serve increased globalization and a sophistication of financial markets, that shareholders' interests have become the focal point for all critical corporate activities. Successful

companies of the future will be those that make managing and creating shareholdervalue (SHV) the central goal of their corporate and businessstrategies. There is no lack of evidence that focusing on optimizing shareholder value is the best way to ensure a firm's long-ten-n prosperity. McTaggart et al. (1994, p. 10), for argued that "maximizing shareholder value is superior to any other governing example, becauseit will lead managersto make the decisions most objective a company might adopt likely to increase the company's competitive, organizational, and financial strength over time". However, corporate managersmay not always engagein transactionsthat are solely in the best interest of shareholders.Studies show that those managerswho fail to deliver to shareholders in the race for global capital resources,will find their companiesat a value disadvantage. As Young and O'Byrne (200 1, p. 13) aptly put it, "They must competitive learn to navigate the rough seas of competitive capital markets, or they will find themselvesreplaced by managerswho can". Rappaport (1998) also points out: "The threat of takeover is an essential means of constraining corporate managers who might choose to pursue personal goals at the expense of Any significant exploitation of shareholders should be shareholders. in a lower stock price. This lower price, relative to what it might reflected be with more efficient management, offers an attractive takeover for another company, which in many cases will replace opportunity " incumbent management. (p.4) Likewise, Copeland et al. (1996) reported that maximizing shareholder value be closely linked with a higher standard of living, greater overall productivity appearsto better functioning equity market. and competitiveness, and a "If countries whose economic systems are not based on maximizing investors lower return on capital than those who do, shareholder value give

they will slowly be starved for capital, as capital markets continue to falling farther and farther behind in global competition. - (p.4) globalize, They go on: "If suppliers of capital do not receive a fair return to compensatethem for the risk they are taking, they will move their capital across national borders in search of better returns. If they are prohibited by law from moving their they will consume more and invest less. Either way, nations who capital, don't provide global investors with adequatereturns on invested capital are doomed to fall farther behind in the race for global competitiveness and decreasingstandardof living. " (p.27) suffer a stagnating or

Empirical studies and business reports show that there has been widespread interest in SHV-based systems and performance measurement approaches worldwide (Ittner & Larcker, 1998; Black et al., 2001). A good number of relatively high-profile have already taken steps towards installing and implementing a SHV corporations in the USA, Europe, and Japan, but also in many other measurementsystem -- not only that of Singapore, South Korea, India, China, Brazil, and emerging economies such as Hungary. To quote Black et al. (2001, p.255) over 5% of the FT 500 companies and about 8% of the FT Global 500 have now installed a SHV/performance measurement system in many cases,the large corporations have beenthe ones to take the lead. and, The growing predominance of the SHV culture is largely a consequenceof several developments,which, among others, major 3 include the following:

1) The globalization and deregulation of financial markets; 2) New advancesin information technology including the internet; 3) Generational changesin attitudes toward savings and investment;
details, seefor example,Shiller (2000); Soros(2000); and Young and O'Byrne (2001). For more

10

4) The expansionof institutional investment; and 5) 'Irrational exuberance

94 .

The global economy is increasingly characterizedby a freer trade in factors as well in goods and services. Among other things, this means that investors now have the as their money much faster and more easily around the world as they possibility of moving in search of the greatest return. As a result, interest rates, exchange rates, are constantly in various countries are invariably interrelated, says American billionaire and stock prices financial mogul, George Soros. He goes further to assertthat global financial markets and tremendous amount of influence on economic conditions throughout the world. In exert a today's business environment, financial capital enjoys a privileged position as it is more than other factors of production (Soros, 2000). mobile Moreover, interest in all kinds of stock: in high tech companies, in new and oldfirms, as well as in general investing, has grown in unprecedentedways in recent economy history. This culminates in a colossal worldwide explosion of mutual funds, unit trusts, and forms of institutional investment like hedge funds. According to some observers, novel have a financial stake in companies, especially, through mutual many more people can funds or pension funds. What is of particular importance to corporate managers,though, is funds are controlled by professional managerswho care only about performance that these delivering the highest possible returns to the people who hired them (Young and and about O'Byme, 2001, p-7).
4 The term "irrational exuberance" was coined by Alan Greenspan,chairman of the US Federal Reserve describe the dot.com boom. It was also the title of a book by ProfessorRobert Shiller from Yale Board, to the boom. In his book, Shiller (2000) arguedthat, as the bull market developed,it University that examined the ftiture and stimulateddemandfor shares. optimism about generated

On the other hand, one could arguethat corporatemanagers often get confused;if
not trapped, when they face multiple objectives or when they are held accountableto more than one party/objective. To avoid such confusion or conflicting signals, they have to focus on only one prime overriding objective. This simplifies matters and makes a great deal of difference. As Brittan (1996) pointed out, "People function best if they have specific for which they are held accountable by means which are transparent, responsibilities verifiable and respect the realities of human nature." The objective of maximizing shareholder value seems to be absolutely sound as it not only allows for just such a but it also provides the transparent and verifiable means necessary to responsibility, measure it. In short, one can conclude that it has all the characteristics of the "right"

objective.
The theories underlying SHV have a long history stretching back to the intellectual work of Markovitz (1952), Modigliani and Miller (196 1), Lintner (1965), Sharpe (1964), Fama (1965) to name just a few. According to Black et al. (2001, p.21), the SHV and discipline started to take on a life of its own as a result of work done on the Capital Asset Pricing Model (CAPM). The fundamental idea behind this model is that the expected is linearly related to its systematic or market risk, as measuredby beta. return on equity The higher the beta, the greater the expectedreturn. Black et al. further point out that: "the key insight of the CAPM model - one that is central to the SHV view of the is that there is a risk-weighted discount factor which allows [one] to assess world the value today of tomorrow's developments, profits and cash flows. This discount
5It is important to note that, as great a developmentas CAPM was, it is not without seriouscriticism. The trying to empirically validate it is so large that we will not attemptto cataloguethem all. numberof papers For more detail, seefor example,Blum and Friend (1973); Famaand French(1992,2004); and Bornholt (2007).
5

12

is derived from observations of capital markets, and defines rate what the opportunity cost of equity to investor in the market is. It stateswhat the company has to earn in order to justify the use of the capital resourcestied in the up business. 22) "(p.

The SHV approach has gained widespread acceptance since the publication of 'Creating Shareholder Valueby Alfred Rappaport in 1986.6This text provided a new and in-depth assessmentof the rationale for the SHV approach as well as the tools neededto implement it as a standard for businessperformance.According to Rappaport (1998, p.32), the total value of an entity such as a firm or businessunit is equal to the sum of the values its equity and its debt. This economic or strategic value of the business is termed of 44 corporate value" and the value of the equity portion is termed "shareholder value". The the firm can then be written as: value of

Corporate Value = Shareholder Value+ Debt Value

(2.1)

Thus, in order to determine shareholdervalue, one must first determine the value of the total firm or business unit, that is, corporate value. To value a company, several writers firms (e.g., Damodaran, 2006; Copeland, 1996; Titman and and management consulting Martin, 2008; McKinsey & Company, 2005), have proposed analyzing the company's historical performance; defining and projecting free cash flow over the short, medium and long run; and discounting the projected free cash flows at an appropriate cost of capital. Using the free cash flow (FCF) approach, the total businessvalue is determined by the so-

A revised and updatedversion of the original edition waspublished in 1998.

13

called "Free Cash Flows to the Firm" (FCFF), discounted at the "Weighted Average Cost of Capital" (WACC). Since the expected FCFF cannot be estimated forever, it is suggested

that it be estimatedduring the "forecast period" of five or ten years and that a "residual
(or terminal value) be estimated for the period beyond the forecast period. The value" present value of the FCFF over the forecast period plus the present value of the residual would result in the value of the businessas a whole. In general terms, the value of a value firm that expects to sustain extraordinary growth for N years can be written as:

Corporate Value

]+ [Terminal ValueN Expected Cash Flow to the Firmt C)N (I + WAC (I + WACC)t t=1

(2.2)

For a more precise estimate of corporate value, a third component to be added to the corporate value model; that is, the current value of marketable securities and other noninvestments that can be converted to cash, which are not essential operating assetssuch as the business (Rappaport,1998, p.33). Thus, the value of a firm can be written to operating as the sum of three components:

CorporateValue = Presentvalue of cashflow from operationsduring the forecast period Residual value (2.3) Marketable securitiesand other nonoperatingassets

After the value of the firm as a whole has been determined, the part of the value is calculated as follows: available to the shareholders

14

ShareholderValue= Corporate Value- Debt Value or

N

SH VO

FCFF, (I+ WACC,)'

[WACCI FCFF N+l

-gN_--

oo CF, (I+ RFI, 1=1

(2.4)

where SHVo FCFFt N

9N

WACCt
CFFI, t

RFI,

= shareholder value in year 0 (current year) free cash flows to the firm in year t = expected = number of years of high or extraordinary growth beyond year N = stable growth = weighted average cost of capital in year t flow of a fixed income security (debt obligations and other = expected cash claims such as preferred stock in year t to be used to discount the cash flows in year t = required rate of return

The idea of measuring shareholdervalue by comparing cash flows generatedby the business against the cost of capital used in generating those flows is to provide a clear degradation over time within each business unit. understanding of value creation or Rappaport (1998, pp.55-57) indicated that shareholder value is driven by seven factors: income tax rate, working capital investment, fixed sales growth, operating profit margin, investment, weighted averagecost of capital, and value growth duration. The theory capital is that improvement in these value drivers leads directly to an increasein shareholdervalue as shown in Figure 1. In the 1990s, interest in the SHV approach received a further boost first by the 'Valuation. - Measuring and Managing the Value of Companies' by Tom publication of

15

Copeland, Tim Koller, and Jack Murrin from the McKinsey GroUP7 and second by the ,

publication of 'The Quest for

8 Value' by G. Bennett Stewart. The Copeland text

demonstratesin great detail how businessescreate value and argues that companiesthrive when they create real economic value for their shareholders.This text further assertsthat by investing capital at rates of return that exceed their cost of companies create value capital (WACC). Copeland et al. (1996) put forward the idea that the application of SHV to companies is both feasible and highly desirable. They also debate whether principles such an approach can yield substantial benefits not only to shareholders,but also to other 'Stakeholders' in a company.

On the other hand, Stewart's book (The Questfor Value) introducesthe idea of
economic value added (EVA) -- a revolutionary new concept that has been developed by the US consultants Stem Stewart & Company to identify and track sources of value that are not explained by traditional accounting and financial measures.While the creation be creative, EVA is simply a variant of the well-known concept called acronym may 9 It is income. the adjusted after-tax operating income minus a capital residual simply Although the term EVA appearedin financial literature as early as 1989 (Finegan, charge. 1989), it did not attract that much attention until an article appearedin Fortune magazine September20,1993.10 Nevertheless, Stewart's text was instrumental in promoting and on advancing this new/old value-basedmetric.

7 Copelandet al. (1996). 8 Stewart(199 1). 9 It is also called economic profit (EP). See,for example,Arnold (2005, p. 828). 10 Tully (1993).

16

It goeswithout saying that all thesebooks have contributedin one way or another
to the promotion as well as to the popularizing of the SHV discipline. By so doing, companies have achieved a real and sustainable increase in their share value. The TM FinanceAdviseorTM 2.0, CAPI emergence of recent software (e.g., EVManager , Balanced Scorecard EVA Model, ART-EVATm) has allowed SHV to continue to advance as more and more companies, which previously had neither considered it nor felt competent or comfortable applying it, are now beginning to implement it. It has long been argued that corporations should be run in order to maximise shareholder wealth. Black et al. (2001, p.257) stated that "The emergence of the SHV concept reinforces the messagethat companies have to improve their returns on invested capital as well as reduce their cost of capital." Friedman (1970), for example, affirmed this by suggesting that the firm's sole purpose should be to operate for shareholders.However, managing a company for value requires delivering a maximum return to the equity holders while balancing the interests of the other important constituents, including customers, employees, government, and suppliers. The proponents of the shareholderwealth approach frequently argue that maximizing shareholder value leads to the maximization of all The argument here is that by adopting the measures necessary to stakeholder claims. a company can advance the interest of other stakeholders as maximize corporate value, its shareholders. This also adds value to the society in which it operates. In the well as literature, this phenomenon has been linked to a win-win situation (Cooper, 2000, p.81). When corporations correctly implement this strategy, not only do the shareholdersbenefit,

17

but everybody else does as well. The following quotation sums it up (Copeland et al., 1996): "Empirical evidence indicates that increasing shareholder value does not the long-run interests of other stakeholders. Winning conflict with companies seem to create relatively greater value for all stakeholders: customers, labor, the government (via taxes paid), and suppliers of capital. Yet, there are additional reasons-more conceptual in nature, but equally compelling-to adopt a system that emphasizes shareholder value. First, is the best metric for performance that we know. Second,shareholders value are the only stakeholders of a corporation who simultaneously maximize in seeking to maximize their own. And finally, companies everyone's claim that do not perform will find that capital flows toward their "(p. 22) competitors. Similarly, in 1996, the CEO of Coca-Cola, the late Roberto C. Goizueta, argued the case for putting shareholders(or owners) interests first: "Saying that we work for our share owners may sound simplistic - but we frequently see companies that have forgotten the reason they exist. They try in vain to be all things to all people and serve many mastersin may even different ways. In any event, they miss their primary calling, which is many to stick to the business of creating value for their owners... [While] a healthy company can have a positive and seemingly infinite impact on is a drag on the social order of things. It cannot others, a sick company jobs, much less widen the opportunities available to its employees.It sustain It cannot give to philanthropic causes. cannot serve customers. ... The real and lasting benefits we create don't come because we do good deeds, but because we do good work - work focused on our mission of I time for the people who own the company."' creating value over finally be concluded that the idea of a firm's operating to maximize It can is not a new one, but this doctrine is only now gaining widespread shareholder value Rappaport (1998, pp. 1-3) concedes that this is now being embraced as the acceptance.
" Remarksdelivered to Executives' Club of Chicago,quotedin Coca-ColaCompany annualreport, 1996.

18

"politically correct" stance by corporate board membersand top managementin the United Kingdom, continental Europe, Australia, and even in Japan.However, in the United States, it has been a long established tradition. Rappoport also assertedthat, as is the case with other good ideas, shareholder value has moved from being ignored to being rejected and then to becoming self-evident. Furthermore, he predicts that over the next ten years, shareholdervalue will more than likely become the global standard for measuring business performance.

2.3 VALUE-BASED

METRICS

Recent surveys have indicted the increasing importance of value-based metrics as benchmarks for assessingand managing corporate performance (see, for example, Ittner Larcker, 1998; Black et al., 2001). In the past 10 to 15 years, many consulting firms and have been caught up in a fierce competition to promote their service regarding value-based They desperately try to capture the hearts, minds, and dollars of performance measures. In an interesting article in the CFO magazine, Myers (1996) has corporate executives. dubbed this engagement as the "Metric Wars". Measures such as Economic Value Added (EVA), Market Value Added (MVA), ShareholderValue Added (SVA), Economic Profit (EP), Cash Flow Return on Investment (CFROI), Tobin's Q or Market to Book Value, Total Shareholder Return (TSR), and other value-basedmetrics are virtually all "rooted in the concept that companies should not look at reported earnings, which are subject to distortions, but at how a company's returns exceed its cost of capital"(Myers, accounting 1996, p.42). Money managers, business analysts, corporate executives, consultants, and

19

academicshave increasingly utilized these metrics as they provide unique advantagesover traditional accounting-based metrics such as EPS, ROA, ROE, or balance sheet ratios in reflecting value creation. In this section, six of the more popular value-basedperformance

discussed. metricsarepresented and

2.3.1 Economic Value Added (EVA) More recently, Economic Value Added (or EVA) has been attracting considerable in the financial press and corporate world. Biddle et al. (1997, p.302), for attention instance,pointed out that citations of EVA in the businesspress have grown exponentially, from I in 1989 to 294 in 1996. Fortune magazine in its breaking article (Tully, rising 1993), branded EVA as "the real key to creating wealth". In August 1997, the Economist interesting article about EVA, crowning it as "a star to sail by". As a also published an new management tool to gauge corporate performance and value creation, it has been broadly accepted by a wide range of senior executives, financial analysts, and institutional investors. Moreover, an AICPA workshop on the future of financial management(April 1995) predicted that EVA would replace EPS (earnings per share) in The Wall Street Journal's regular stock and earnings report (Zarowin, 1995, p.48). Today, in North America and around the world, a sizeable number of companies have adopted EVA as their key performance metric, even linking it to the fortunes of their The growing popularity of EVA has also been reflected on the capital and executives. increasing number of security analysts at brokerage housesare money markets, where an EVA to pinpoint winners and losers (Topkis, 1996). Furthermore, as companies and using

20

for their officers are increasingly being held accountable shareholder value, EVA, which forces managersto think and behave like shareholders, becoming more and more a is fundamentalpart of both running andjudging a business.
The concept of EVA is neither new nor complicated. It dates back to Alfred Marshall, a famous English economist who, over 100 years ago, wrote that a firm, in order to create real earnings, must generatea profit in excessof its capital cost (Marshall, 1890). What Marshall was actually saying is quite simple. A firm can only createtrue value for its if and only if it is capableof making sound investment, financial, and shareholders/owners, decisions, which yield a return in excess of its cost of capital. If that is not the operating then there is no real profit or true value added to the businessand, actually, from the case; the company would thus be operating at a loss. This valuable shareholders' viewpoint, insight of Marshall's has been a good management practice for business growth and the beginning of last century. survival since It is interesting, though, to note that, the managementguru, Peter Drucker, in the fifties as well as in a more recent 1995 Harvard Business Review article, reiterated this idea by saying that "Until a business returns a profit that is greater than its cost of capital, it operatesat loss. Never mind it pays taxes as if it had a genuine profit. The enterprise still a less to the economy than it devours in resources. It does not cover its full returns the cost of capital. Until then, it does not costs unless the reported profit exceeds it destroys it. " (Drucker, 1995, p. 59). create wealth; In the twentieth century, the concept of EVA was the object of extensive academic debate in the accounting and finance literature and it has been operationalized under

21

labels including Residual Income (RI). 12 In the past several years, a US-based various business consulting firm Stem Stewart & Company (hereafter referred to as Stem Stewart) has been promoting a variant of RI under the acronym of Economic Value Added or EVA for measuring corporate performance and value creation.13This new/old measure, as a tool though, is similar to residual income (RI), but distinguishes itself by a series of to eliminate potential distortions of accrual accounting as well as the inclusion adjustments 14 both debt and equity sources of capital in the calculation of cost of capital. of According to Stem Stewart, EVA is defined as the difference between a firm's net income after taxes (NOPAT) and an appropriate charge for the opportunity cost operating invested in that firm (Stewart, 1991, pp. 136-138). As such, EVA is a measure of all capital to produce an economic profit, that is, a return in excessof a firm's of a company's ability In equation form, EVA for a given year, t, can be expressedas follows: cost of capital.

[WACC, x TIC, ]= (ROIC, WACC, ) x TIC, E VAI = NOPA T, -, -,

(2.5)

NOPATt is the net operating profit after taxes, but before financing costs in year t; where TICt-I is the economic book value of the total capital invested in the company, at the
12 Residualincome (RI) is generally defined as what is left from accountingearningsafter deductinga charge for investedcapital to reflect a minimum required rate of return on the investedcapital. Book value is usedas investedcapital. Thus, a firm's RI for any period is calculatedas a product of the the measureof betweenthe firm's return-on-equityand its equity-cost-of-capitalmultiplied by the firm's 'spread' beginning of the period. It is interesting,though, to note that GeneralMotors accountingequity value at the in the 1920sand GeneralElectric coined the term 'residual income' in the applied a variant of this concept divisions. See,for example,Bromwich and it to assess performanceof its decentralized the 1950sand used Walker (1998, p.392). 13 developedin 1982by the Stern Stewart& Company.SeeGrant (2003, p. 1). EVA was commercially "' See,for example,Madden (1999, p.202).

22

beginning of year t, which includes both the interest bearing debt and equity'5 it stands -as proxy for all cash invested in the company since its inception; WACCt is the weighted average cost of capital in year t, that is, the minimum rate of return demanded by both lenders and shareholders; ROICt is the return on the capital employed in the company in t and is calculated by dividing NOPATt by TICt-1; [ROICt - WACCJ is the year [WACCt * TICt-1] is the annual capital charge, i. e., the cash flow profitability spread; and required to compensate all the company's capital providers, equity as well as debt, for the the capital that has been used during the year. risk of Thus, for an ongoing concern, a firm's EVA can be defined as the difference between its un. levered net operating profit after tax (NOPAT) and a dollar charge for the in the business-as measured by the amount of total invested capital capital employed (TIC) times the weighted average cost of capital (WACC). The NOPAT in the EVA model the total pool of profits available to provide a cash return to all financial represents to the firm. It can be expressedin terms of the firm's pre-tax operating providers of capital EBIT, less unlevered operating taxes (T): profit,

NOPA T= EBIT(I - T) = (S - COGS- SG &A Exp - Dep)(I - T)

(2.6)

As is shown in equation (2.6), EBIT is a function of the firm's sales (S) less expenses including cost of goods sold (CoGS); selling, general and administrative expenses(SG&A Exp); and depreciation (Dep). Unlike the operating profits calculated by many companies,
15It is important to note that Stern Stewart make a number of adjustments for their publicly available database as well as for their corporate clients not all of them available to the public.

23

both depreciationand businesstaxes (T) are subtractedfrom NOPAT because they are
genuine economic costs that have to be managed(Ehrbar, 1998, p. 131). In turn, for a firm financed solely with debt and equity, the weighted averagecost (WACC), is defined as follows: of capital

WACC = (WD)(RD)(1- T) + (WE)(RE)

(2.7)

where,
WD= the proportion of total market value (debt plus equity) contributed by debt capital Wx, the proportion of total market value (debt plus equity) contributed by equity capital = T =Tax rate RD= pre-tax cost of debt
RE: -'-'cost of

equity

16is defined as the sum of the working capital Total invested capital (TIC) usually fixed assets. The working capital requirement consists of accounts requirement and net inventories, net of accounts payable and accrued expenses. Taken together, receivable, these financial developments show that the firm's EVA can be expressedin basic terms as

follows:

[S CoGS SG &A Exp Dep], (I T) [WACC, x TIC, ] EVA, = _,

(2.8)

16 Also called "capital" or "capital employed".

24

The aboveEVA model (Eq. 2.8) demonstrates EVA is a financial management that
that integrates operating efficiency and balance sheet management into one easily system that can be understood by all managers and operating people. The accessible measure that EVA is a performance metric that takes into account the cost of the model also shows the company employed -- a factor that no conventional measuresuch as accounting capital earnings per share (EPS), return on assets (ROA), or return on equity (ROE) earnings, includes. As a matter of fact, the opportunity cost of capital is what makes EVA a truly business value. 17By focusing on the financial metric and an accurate gauge of unique the opportunity cost for all capital employed, EVA profit remaining, after subtracting the real return a company can get on investor dollars. With EVA technology in shows hand, investors have the upper hand. If, for whatever reason, a firm cannot generate to cover its cost of capital; its investors will reduce the flow of capital to that enough return drastically re-price its stock downward to reflect their lower company and more expectations. The model also reveals that EVA is not only a measure of performance, but it is It measureshow and if a company createstrue value for also a measure of value creation. its shareholders. A company with a positive EVA creates value; a zero EVA maintains EVA suggestsa squanderingof value. In symbols, value; whereas a negative

17 should be noted, however, that like EVA, the accountingmetric ResidualIncome (RI), makesa cost of It over and abovethat the firm's profit to calculatewhat value is being generated equity capital chargeagainst investors.However, RI is much lesssophisticated to mention lesspopular than EVA due not requiredby the by consultantsas well asdue to the fact that it lacks the necessary adjustments to the absence support of balancesheetand the income statementfigures to removethe accountinganomalies.See equiredto both the for example,Francis and Minchington (2000).

25

If [ROIC - WACC] x capital employed >04 If [ROIC - WACC] x capital employed =04

value is created is maintained value

If [ROIC - WACC] x capital employed <0 --) value is destroyed

Stem Stewart and EVA proponents argue that EVA is much more than just a It is a modem management system that challenges a company to look corporate metric. hard at how it conducts and governs its business.Ehrbar (1998, p.5), for example, argues that EVA is definitely not another form of rightsizing or downsizing, nor is it a fad. It is a "fundamental way for measuring and managing corporate performance... It tells managers to do those things that they intuitively know are the right things to do, but that so often are obscuredby conventional accounting-basedmeasuresof performance." The appeal of EVA lies in its ability to integrate the often disparate management functions of strategic thinking, measuring performance, evaluating new investment decisions, communicating with opportunities, operating investors, and motivating

When EVA becomes the singular focus for all decisions, it establishesclear employees. links among all corporate functions. CFO Basil Anderson of Scott Paper and accountable states:
18

"We used to have different financial measuresfor different purposes- discounted flow for capital decisions, another measurefor rewarding performance and the cash like. Now EVA is one measure that integrates all that. it offers an excellent ... ... link to the creation of shareholdervalue."

18 Cited by Walbert ( 1995, pp. II 1-112).

26

Jim Meenan, CFO of AT&T's communications services group expressesa similar view:

"Every decision is now basedon EVA. The motivation of our business is no units longer just to make a profit. The drive is to earn the cost of capital. when you ... drive your businessunits toward EVA, you're really driving the correlationwith the market value."(Walbert, 1995,p. 112)

In fact, EVA is the only financial management system that provides a common language across all corporate functions. No other financial measure can do this. Figure 2 illustrates a typical financial management system where multiple measures and

terminology are used to express financial goals and objectives. Usually, the end results of such a practice are confusion, inconsistent standards,and above all non-cohesive decision making. In contrast to this traditional system, Figure 3 shows how EVA can be used as a managerial tool to streamline and simplify the whole managementprocess; thus providing a more consistent framework for decision-making. Needless to say, this type of is of utmost importance as it unites and aligns the interests of all corporate consistency employees-from CEO to CFO, to factory managers,to floor operators-with the overall goal of increasing shareholder wealth. The most obvious way that EVA helps managers make better decisions for their is by charging their operations for the cost of all capital employed. The capital companies is, in fact, what sensitizes managersthroughout a company to pay closer attention charge to the investment on hand. Indeed, EVA compels employees to use assetsmore diligently One company's experience with EVA, for instance, was summed up by and efficiently. Federal-Mogul's CEO Dick Snell in the statement:

27

"EVA accelerated ability to divest assets focus on our manufacturing our and core, to restore earningsimprovements,and to begin to grow by acquisition.EVA was the criterion we used to evaluate each and every action-whether for improvement,or growth. With the clarity it provided, rationalization, continuous EVA allowed us to complete our evaluationsquickly and move on to the next challenge.We are confident that EVA will continueto help all of us to makebetter businessdecisions,to build a world-class company, and to enhanceshareholder "19 wealth.

In short, using a single financial tool, such as EVA, not only eliminates the type of that CEO's and other senior executives are concerned about; but most confusion importantly, it links all decision making to a common focus: how to improve EVA, and how to make companies more valuable. Furthermore, as many CEO's are discovering, EVA can also be used to transform from top to bottom. For example, the CEO of Quaker Oats, William corporate culture, Smithburg who supported this notion, cited that "Our operating managersare much happier that we have a single measurethat embodies all of the things we want them to think now in running their business. Before, we used a whole variety of measures and about for different purposes, and that only servedto confuse our managersrather than procedures (Stewart, 1994, p.74). In this regard, EVA could spearheada to clarify their mission" in management as it better answers the problems of how to align the genuine revolution interest of agents with principals and of how to bind managersand employeesto the will of the shareholders(Ehrbar, 1998).

19Cited by Ehrbar (1999), p.221.

28

Obviously, EVA isn't without some problems. In an attempt to allow EVA to approximate value maximization, Stem Stewart recommends making numerous

to the accounting assets and NOPAT. These adjustments are necessaryto adjustments introduced by GAAP, as well as to make EVA a more eliminate any accounting anomalies To date, Stem accurate period-to-period measure of performance and value creation. Stewart has identified and catalogued a total of 160 potential adjustments and they are 20 designed to reflect three purposes: primarily a) To undo accounting conservatism; To immunize performance measurementagainst past accounting "errors";

and
C) To make current EVA a better measureof market value.

Some of the more common adjustments proposed by Stem Stewart are listed

21 below:
9 9 0 o 0 e e Capitalize R&D expenses. Capitalize operating leaseexpenses. Add back any inventory LIFO reserves. Add back deferred tax reserves. Add back bad debt reserves. Add back one-time restructuring charges. Add back amortization of goodwill.

20For a completediscussionincluding somenumerical examples,seeYoung and O'Byrne (2001, chapter6); (2003, chapter9); O'Hanlon and Peasnell(1998); and Ehrbar (1998, pp.161-181). Grant
21

SeeStewart, 1991, p. 112.

29

While it is conceptually sound to adjust for various distortions as proposedby Stem Stewart, it has been correctly argued that most of these adjustments are fairly complicated

and insignificant, some of them are even inconsistent with others,and aboveall many of
them often call for excessive reliance on judgement'. As aptly put by O'Hanlon and Peasnell (1998, p. 442), "Stem Stewart appears to have devised the accounting in an essentially ad hoc fashion, on the basis of consulting experiences,and adjustments, the adjustments do not seem to be clearly underpinned by any formally expressedtheory of income measurement or some would put it more bluntly, arbitrary decisions. " At the

Whirlpool Corporation, for instance, the merit of 160 adjustments was debated prior to implementation, and a decision was reachedto make the adjustments only for factors that big difference. In the end, adjustments were made in only four areas: goodwill, made a leases, restructuring charges - which are each viewed as investment and capitalized interests in investments.22 minority Apart from any further accounting difficulties that might arise when estimating NOPAT and TIC, in practice, many "cost of capital" issues remain debatable that can impact the estimation of EVA. These cost of capital challenges have both theoretical and foundations that need to be tackled and resolved (e.g., Fabozzi and Grant, 2003). empirical While estimating the cost of debt is fairly straightforward due to the fact that bond yields the measurementof the cost of equity poses estimation difficulties are readily observable; from the uncertainty surrounding the appropriate model and potential estimation arising is used. Despite all its limitations, recent reports have shown that errors whichever model
22Cited by Shaked(1997, p. 3).

30

capital asset pricing model (CAPM) remains the most commonly used approach to firm's cost of equity.23 The Sharpe-Lintner CAPM risk-retum relation to estimate a estimate the cost of equity capital is an expectational model and its basic intuition can be

follows: as summarized
E(R, ) = Rf +, 6, [E(R,,, ) - Rf ] (2.9)

E(R) denotes the expected return on asseti, Rf is the expected return on a risk-free where (such as a government bond), 6, = cov(R,, R )/ var(R,,, is a measure of volatility ) asset ... it representsthe sensitivity of the asset'sreturn to variation in the market return. E(R,,, ) and is the long-term expected return on the stock market (e.g., S&P 500, FTSE-100, or some index). Thus, the expected return on a risky asset, such as an equity other market investment, equals the return on a riskless assetplus a risk premium. That risk premium be obtained by multiplying the market risk premium [defined as the difference between can 8,. [E(R. ) and Rf ] by the asset's market beta,, Suffice it to say, that determining and CAPM's components is crucial to the estimate of (Ri) and properly calculating each of deal of judgment and interpretation which might somehow lead to differing requires a good (Young and O'Byrne, 2001). It is also important to note that in the CAPM, the conclusions to all firms; only beta varies across risk-free rate and market risk premium are common

firms.
While CAPM is a widely used model for estimating the cost of equity, several Fama and French (1992,1993,1995,1996,2004) empirical studies, most notably
23See,for example,Bruner et al. (1998), Famaand French(2004), Rutterford (2000), and Grahamand Harvey (2001).

have

31

raiseddoubtsabout the validity of the single-factorCAPM model as an accurate estimator

the cost of equity. Fortunately, financial economists and consultants have devised and of that can be used to estimate a firm's cost of equity proposed more sophisticated models in lieu of the single-factor CAPM. Nevertheless, in the absenceof other simplified capital CAPM provides a relatively good starting point for estimating the cost of equity models, (Perold, 2004; Palliam, 2006). Furthermore, and as noted earlier, calculating EVA could provide a positive or dollar value, which indicates whether the firm earned an excessabove its cost of negative during any given year. Critics of EVA claim that there are two problems with this capital dollar value of EVA (Reilly and Brown, 1997, p.741): First, how does one judge, annual time, if the firm is prospering relative to its past performance? Although you would over the absolute EVA to grow over time, the question is whether the rate of growth of want EVA is adequate for additional capital provided. Second, how does one compare EVA business units of different sizes? To rectify these problems, Stem among companies or Stewart proposed that EVA be standardized or normalized by dividing it by the capital the beginning of the evaluation year, and multiplying by 100 (Stewart outstanding at (1991), p. 167). This effectively expressesEVA as a percentage of the beginning of year capital:

ITICI-1 STANDARDIZED EVA, =]=

(NOPAT, Ticl-l WACC, x 100 (2.10)

32

Clearly, one would want this EVA rate of return on capital to remain constantover time, or ideally to grow. Also, using this ratio one can compare firms of different sizes to determinewhich one has the largest economic profit per dollar of capital. This study usesthe standardizedEVA, that is, the annual EVA deflated by the beginning-of-year capital as the

independent variable.
Last but not least, and despite theoretical arguments and some empirical evidence favouring the adoption of EVA worldwide, the implementation of this fairly sophisticated financial management system is less straightforward than one based on traditional 24Moreover, implementation of the EVA system a successful accounting-basedmeasures. long-term commitment from top management, and intensive requires, among others, training programs for employees. In an article published by Fortune magazine (1995), G. Bennett Stewart 111, the originators of EVA and a senior partner at New York-based one of Stem Stewart & Company, refuted critics of EVA and argued that some of the common in implementing the EVA measure may lead to the failure of EVA (Stewart, mistakes 1995). He concisely outlined five pitfalls that companies may fall into when implementing EVA. If unaware of these traps; companies will pay the consequences.These are the following: 1) Companies not making EVA a we of liLe:

it 24It is interestingto note that in 1992AT&T adoptedEVA but subsequently abandoned for variety of for the EVA's demiseare the complexity of the metric and the difficulty of Someof the reasons reasons. the impact of their actionson the EVA metric combinedwith the hiring of a communicatingto employees had not championedthe EVA system.For a discussionon AT&T's experiencewith EVA, see new CEO who Ittrier & Larckre (1998, pp.215-217).

33

Calculating EVA is probably an important step, but it is certainly not sufficient. EVA has to be the focus of a company's financial managementsystem and be fully integrated into every managementdecision. EVA should never be a supplement to the reporting process,a line on the page, or one measure in a "balanced scorecard". It must be the bedrock upon the entire reporting process is built. EVA must also constantly be reinforced in which managementmeetings, training seminars, company newsletters, performance reviews, and in communications with external parties such as security analysts and the financial press. 2) Companies jing to implement EVA too fi7s : Becoming an EVA company, simply, cannot be done overnight. The larger the company, the longer it will take to fully implement it. The changeover has to start with top its way down through the ranks and it may take months management,and gradually work or even a couple of years. 3) Manazers not being sure: Overpaid, mediocre or incompetent managers,who have an incentive plan that works for the implementation of EVA. What incentives are there for them to switch them, often resist EVA system, under which they will suddenly be held accountable for creating to the Thus, compensation and bonus determinations have to be tied to changesto EVA as value? to make it count. a way 4) Managers creating too much Luss. EVA, many companies tend to make it into a big philosophical issue When implementing instead of keeping it simple. Managers get distracted by endless rounds of worthless discussion about what it means to create value for the shareholder as opposed to the

34

the community at large. The bottom line of EVA is that it is beneficial for all employee or the company's stakeholders, once it has been well integrated into the decision making process. 5) Not enough training: Some companies do not widely disseminate EVA knowledge throughout the organization. To fully benefit from EVA, it is important to intensively train everyone in the organization because even those with the smallest jobs can help create value. Employees must how their unit EVAs will be calculated and how, in detail, the EVA numbers understand be linked to compensation. will

2.3.2 MARKET VALUE ADDED (MVA)

EVA is designed to be a single-period measurethat tells what has happenedto the wealth of shareholders. For publicly traded companies, Stem Stewart developed another that indicates whether or not a company has created an additional value to its measure shareholders'wealth. This cousin of EVA was termed 'market value added' or MVA in is defined as the absolute dollars spread between the current market value of the short and firm and the total capital that investors have committed to it since its formation (Stewart, 1991). In other words, MVA is the difference between cash in (what investors have the inception of the company) and cash out (what they could sell their contributed since for today). So, if a company, for instance, has a total market value of $10 million, claims if it had invested only $8 million capital, then that company would have a $2 million and

35

Market Value Added. If this samecompany,though,had invested$11 million capital,then

its Market Value Added is a negative $1 million. In calculating MVA, Stem Stewart first adds up all the capital taken in by a during its lifetime through securities offerings, bank loans, and retained earnings. company Second, it then makes some EVA-like adjustments (such as capitalizing and amortizing 25These and other adjustments R&D expenditures) to eliminate any accounting anomalies. in order to convert accounting book value into economic book value or what are necessary; Stem Stewart calls 'adjusted book value'. In effect, the adjusted book value is what book be, if accountants were financial economists (Ehrbar, 1998). Then, third, it value would this adjusted capital amount from the current market value of both the company's subtracts 26In equation form, MVA is calculated as follows: debt and its equity. M VA, ": MV company, l " where, MVA
Wcompany
BVassels, l

(2.11)

= market value added = market value of company Stem Stewart's adjusted book value of assets-in-place =

BVassets

25Stem Stewartproposesa seriesof adjustmentsto the NOPAT and the book value with the intention of to EVA and the book value. For more detail, seeStewart(1991). giving more economic meaning 26The cuff ent market value of the firm being valued is basedupon the market's perceptionof the firm's The equity figure is simply the stock price multiplied by the numberof shares future performance. The debt figure is the amountof outstandingdebt owed to lenders. outstanding.

36

Armed with the above definitions and given that the market value of a company
be expressedin terms of present value of all future cash flows and then in terms the can of book-value of assets-in-placeplus the presentvalue of all the EVAs it is current expectedto in the future, the MVA equation can subsequently written as 0 be generate
f
MV, ompany

IIOWS. 27

00
=II=

FCFF,

BVamels-in-place

(I+ WACC) 1=1

Vassels-in-place =

MVA = MVc,,,, -B w,, y

00

EVA, co +EI (I+ WACC) =, EVA,

(2.12)
(2.13)

(I+ WACC)'

where, FCFF WACC MVA

Wcompany

free cash flow to the firm = weighted averagecost of capital = market value added = market value of company

BV.,, t,= Stem Stewart's adjusted book value of assets-in-place

Stem Stewart argues, correctly, that the present value of all future EVAs constitutes the crucial difference between the book value of assetsand the market value of the firm the assets. This excess is referred to as the firm's unrecorded goodwill, or MVA owing (O'Hanion and Peasnell, 1998, p.425). The value creation can then formally be linked to EVA by assessingthe excessof market value over book value as shown in Equation (2.13).

27Seefor exampleO'Hanlon and Peasnell(1998).

37

Thus, maximizing the present value of the future stream of EVA amounts to exactly the thing as maximizing intrinsic market value added. same As such, MVA is greatly influenced by a firm's EVA performance. If the PV of discounted future EVAs turns out to be negative which implies that the firm has not earned enough during the years to cover its cost of capital, then the company's value is expectedto In contrast, if the PV of discounted future EVAs is Positive, then the shrink. company will generate a positive MVA, which will increase the shareholderswealth. The following

the summarizes major calculations:

if

IE

00

VA

(I+ WACC) 1=1 VA IE I (I+ WACC) 1=1

00

0 --) MVA >0 --) is created wealth

if

MVA <04

is destroyed wealth

The interpretation of MVA is quite simple. A positive MVA signifies that a firm has created true wealth for its shareholders,since the company's market value is greater than the book value of the total capital employed in the business.On the other hand, when a firm has a negative MVA, its market value is less than the capital that shareholdersand bondholders invested, meaning that its managers have destroyed capital and squandered shareholder wealth. From the standpoint of assessing the performance of current the change in MVA over a period of one year or five years can be more management, than the absolute level of MVA. Thus, to increasethe spread between invested significant be the primary objective of any company concernedabout capital and market value should

38

(Stewart, 1991). Moreover, as the definitive measureof wealth its shareholders' welfare the ultimate goal in the wealth creation game; MVA could be used to creation and as directly comparethe performanceof companiesin different industriesor even different (Ehrbar, 1998).For instance,MVA could be usedto compare,say,a bank and a countries
toy manufacturer and a food processor or a steel maker and a software supermarketor a

Without a doubt, the one with the higher MVA createdmore wealth for its company.
shareholders. One of the best ways for companiesto build MVA is to generateconsistentlypositive EVA. EVA advocatesargue that companiesthat rack up positive EVA year after year should their MVA soar so their shareholdervalue would rise as well. Companiessuch as Cocasee Cola, GeneralElectric, Microsoft, and Wal-Mart havepositive MVAs because their EVAs are both positive and are generally growing at an exceptional rate over time. In contrast, firms having negative EVA reports should see a noticeable decline in their equity values as the EVA outlook lowers the intrinsic value of the firm (Grant, 1996;Ehrbar, 1998;Grant, adverse 2003).

2.3.3 Tobin's Q (or Market to Book Value (MTBV))

Tobin's Q, a value-basedmetric named for its originator, JamesTobin, a Yale

University professor of Economics, has gained broad acceptanceas a measureof corporate its introduction roughly three decadesago into the financial economics performance since

39

literature.28 Defined as the ratio of the market value of the firm outstanding financial

its tangibleassets, Tobin's Q is often usedasa securitiesto the current replacement cost of
reliable measureof a company's growth opportunities and its ability to create long-run firm value. Its attractiveness results from its ability to provide an estimate of the firm's intangible assets,which include market power, goodwill, future investment opportunities, high quality management; the greater the value of these intangibles, the greater the and

value of Q (Perfect and Wiles, 1994; Tobin, 1969). Thus, changesin Tobin's Q value
important indicator of corporate performance and value creation. provide an A growing number of empirical studies employed Tobin's Q metric to categorize companies according to their relative performance. Lindenberg and Ross (1981), for low Q ratios as competitive, tightly regulated,or in example, characterized companies with dying industries, while companies with high Q ratios tended to have unique products and factors of production. Likewise, firms with aQ ratio greater than one are judged as using

scarce resourceseffectively; whereasthose with aQ ratio of less than one as using

Stated differently, Tobin's Q should be greater than 1.0, if the firm's resources poorly. its investments exceedsits cost of capital. return on Landsman and Shapiro (1995) examined the relationship between Tobin's Q, return

investment(ROI), and economicreturn; andfound that Tobin's Q is a bettermeasure of on Tobin's Q is also firm's economic performancethan other accounting-based measures.
tool for investigating whether a firm is more prone to take over. gaining popularity as a Lang, Stulz, and Walkling (1991) found that firms with low Tobin's Q are more likely to
" For details, seeTobin (1969).

40

be taken over for purposesof restructuringand increasingvalue. In the literature,Tobin's definition may be statedas follows:29

Tobin's Q

market value of the firm R VA

MVCS +L VPS +B VD R VA

(2.14)

The numerator in the equation (2.14) consists of the aggregate year-end market value of common stock (MVCS), the liquidating value of preferred stock (LVPS), and the book debt (BVD). The denominator [the replacement value of assets (RVA)], is value of measured as the dollar outlay needed to purchase the current productive capacity of the firm at minimum cost with the most modem technologies available.

As an alternative to traditional performance metrics, Tobin's Q has several

adherentsin academia but, largely becauseof the unavailability of data, it still has not been to break through into practical use (e.g., Damodaran, 2002, p.538). Moreover, Perfect able and Wiles (1994) maintain that, although Tobin's Q is theoretically an attractive corporate its estimation is subject to considerable measurement error. performance measure, Shepherd (1996), for example, notes that, while the numerator supposedly representsthe the firm, only the market value of the common stock is commonly used. market value of The remaining components are at their book values, or are arrived at using complex and debatable methods. Similarly, the construction of the denominator is controversial and biases. Hence, combining two contains an unknown amount of error as well as potential

29See, for example, Perfect and Wiles (1994).

41

imperfect values results in aQ ratio with the possibility of considerablemeasurement

errors.

While the numerator of Tobin's Q ratio is relatively easy to computeusing data

from databases,the estimation of the replacement cost of assets(the readily accessible denominator) is fairly complicated and, in many cases, hard to obtain due to the data (e.g., Lewellen and Badrinath 1997). In an attempt to deal with this unavailability of data, several estimation techniques (proxies) have been proposed. unavailability of Lindenberg and Ross (1981), for instance, developed one of the most popular ones. They divide the firm's assets into three components, namely plant and equipment, inventories, and other assets,and suggest an appropriate methodology for each. The replacementcost is adjusted to account for four primary effects, including price level of plant and equipment depreciation, technological changes, and investment in a new changes, real economic plant. While Lindenberg and Ross's estimation technique may be theoretically superior, it is still difficult to put into practice. Hence, a simplified version has been adopted by a including Lang and Stulz (1994) and Perfect and Wiles (1994). number of researchers Moreover, in practice, analysts often use shortcuts to arrive at Tobin's Q, using market debt as a proxy for the market value of assetsand book value of assets value of equity and for replacement value. Thus, this proxy for Tobin's Q resemblesthe market-toas a proxy

42

book value (MTBV) ratio and is shown to be empirically closer to the more complex
Lindenberg and 30 Ross proxy:

B VD +L VPS + MVCS for Tobin's Q= proxy B VTA

(2.15)

where, B VD =book value of debt L VPS=liquidation value of preferred stock MVCS = market value of commonstock BVTA = book value of total assets

2.3.4 Total Shareholder Return (TSR) Total shareholder return or TSR is another value-basedmetric which measuresthe for shareholdersover a given period of time. It has been used for many years overall return by investors and business analysts as a meansof assessing performance and value creation. Moreover, TSR target has become an important element in determining executive pay (see Atrill, 2003, p. 366). It is a comprehensive measurethat reflects all activities or decisions taken by a management team and as such it has become an increasingly important indicator (Arnold, 2005, p. 854). Management Today (March 1997, p.48), for of managerial success that "TSR reflects the measure of successclosest to the hearts of a example, pointed out investors: what they have actually gained or lost from investing in one set of company's than in another." executives rather

'OSeeChung and Pruitt (1994); and Perfect and Wiles (1994).

43

TSR showsthe total return shareholders earnedon their shares over a statedperiod time, which in addition to the streamof actual dividend paymentsalso includescapital of
depreciation - any increases (or decreases)in the share price. For oneappreciation or period TSR:
31

TSRI

I DPSI+l + PPSt+l - PPS, x 100 PPS/

(2.16)

where,
TSRI = total shareholder return DPSt+l dividend per share at the end of the period PPS, share price at the end of the period +j PPS, = share price at the beginning of the period (or initial share price) TSR is thus the most direct measureof changesin shareholderwealth over a given period of time, expressed in percentage terms (Young and O'Byrne, 2001, p.417). It has the benefit of being easily understood, and is more "dynamic" in that its values are constantly being assessedby the market. Furthermore, as no accounting information is included in this metric one may claim that it is independentof accounting policies (Whittington, 2000, 360). TSR can be easily compared from company to company without having to worry p. bias. Recognizing this, the "Wall Street Journal" publishes a yearly report about size "Shareholder Scoreboard" that ranks the 1000 major U. S. companieson TSR. called Similar to ROI, TSR, despite its appeal as a measure of shareholder wealth is still considered flawed (Savarese,2000, p.7). The calculated or absolute TSR creation,

" See, for example, Arnold (2002), p. 677.

44

has little information value when taken alone. It does not indicate whether the return
is adequate, that is, is 20% for one particular company better than 15% for achieved A higher TSR does not mean that more shareholder wealth has been another company? This deficiency can be mitigated by comparing the absolute TSR to a benchmark. created, Perhapsthe best benchmark to use is the required return to equity; that is, the opportunity the returns from comparable firms operating in the same industry over the same cost or 32 Thus, to assess period of time. performance, the TSR metric should be used in benchmark to filter out economy-wide or industry-wide factors. A firm conjunction with a for its shareholders if the TSR is greater than the required return to equity creates value (Re), otherwise it destroys value. In symbols,

If TSR > Re --) value is created If TSR = Re 4 value is sustained If TSR < Re 4 value is destroyed

2.3.5 The Cash Flow Return on Investment (CFROI) Cash Flow Return on Investment (CFROI) model is another prominent value-based is consistent with the principles of wealth maximization. 33The model is rooted metric that in the Internal Rate of Retum (IRR) literature and was originally developed by Holt Value

32The reasonthis benchmarkis usually suitable is because comparesthe returnsgenerated the firm with by it thosegeneratedby other investmentopportunitiesthat havethe samelevel of risk. Other benchmarkswould be industry averagesand stock market indices such as S&P 500. 3' For a complete and rigorous coverageof this metric, seeMadden (1999).

45

Associates,a Chicago-basedconsultancy and it is now used by several well-known

firms, including the Boston Consulting Group (BCG), Price Waterhouse consulting Coopers,Deloitte & Touche, and several others (Young and O'Byme, 2001, p.381). According to Madden (1999, p. 13), CFROI is "an estimate of the real rate of return

by a firm on all its assets, " which can be thoughtof as a portfolio of projects. It can earned
be obtained by finding the rate of return that equatesthe present value of the gross future flows available to the firm's debt and equity holders to the gross investment made by cash the capital owners. These cash flows are expressedin real (instead of nominal) terms by for period-to period changes in the general price level. In fact, adjusting for adjusting inflation is one of the distinctive selling features of CFROI as it facilitates comparisons

time and acrosscountries. across

As the term indicates, CFROI is the rate of return earned by the firm's existing Differently put, CFROI is an IRR-type metric, which measuresthe expectedrate projects. 34It is "an IRR measurebut not in of return over the average life of a firm's existing assets. the traditional sense", say Petersonand Peterson(1996, p.26). The CFROI model functions and is normally calculated as an economic, cash-basedmeasure of corporate performance basis. It should then be compared with the inflation-adjusted cost of capital, on an annual COC,35to determine whether a firm has earned returns superior to its cost of capital and thus created value for its shareholders. If CFROI is greater than the cost of capital

34It is worth noting that this analysiscan be done entirely in nominal terms where the internal rate of return is a nominal IRR and therefore is comparedto the nominal cost of capital, or in real terms, in which casethe internal rate of return is a real (inflation-adjusted) IRR comparedto the real cost of capital. 35Inflation-adjusted COC = [(I+ nominal cost of capital) / (I+inflation rate)] -1

46

(sometimes the hurdle rate), wealth is created;otherwise,wealth is destroyed the called if

CFROI has fallen short of the overall cost of debt and equity capital. In symbols,

If CFRO1 > Inflation-adjusted COC 4 value is created If CFROI = Inflation-adjusted COC 4 value is sustained If CFROI < Inflation-adjusted COC 4 value is destroyed

For a particular firm or project, CFROI can be calculated using the following inputs: i)
36

The first input is the gross investment (GI) that the firm has in its existing assets, which is computed by adding back depreciation to the net assetvalue to arrive at an estimate of the original investment in the asset. The gross investment then is converted into a current dollar value by adjusting for inflation.

ii) The second input is the gross cashflow (GCF) earned each year over the expected life of the assets. This is usually calculated by adding non-cash charges such as depreciation and amortization to the after-tax operating income. iii) The third input is the expectedlife of the assets-in-place(n). iv) The non-depreciated assetswhich representthe salvage value (SV) or the expected the assetsat the end of their useful life, in current dollars, is the final input. value of

36SeeDamodaran(2001), pp.453-454.

47

The CFROI is thus the IRR of the above three cash flows and can be computedas follows:
n-

Gross Investment (GI) =I-+ (I + CFROI)' 1=1

GCF,

SV, (I + CFROI)

(2.17)

CFROI is in fact an efficiency measure that compares future flows with the cash total investment employed to generate those cash flows. From the viewpoint of its users, CFROI is considered an informative and fairly useful metric for evaluating the true economic profitability of the firm's existing projects. Its usefulness lies in its ability to become an input into resource allocation decisions since, relative to the cost of capital, it very clearly defines which businesses(projects) are profitable and which are not as well as investment is likely to createvalue. where As CFROI is basedon both current and future cash flows, its advocatesbelieve that the measureis more closely aligned with shareholderreturn. A study by the Chicago-based firm, Holt Value Associates,37showed a significant association between CFROI and stock prices over a 15-year period of analysis. According to this study, there was a 70% factor between CFROI and stock prices, versus 31% for ROA and 44% for correlation ROE. On the other hand, there was a zero correlation between earnings growth and stock Needless to say, Holt is hardly unbiased and the study has yet to be examined by prices. managers. Furthermore, BCG/Holt claims that CFROI avoids the researchersand money

37"Best practice technique: Focus on CFROI analysis to boost your firm's growth in 2004. " 0843&RQT=309&VName=PQ http: //proguest. umi. com/pgdweb? did=466955401&sid=I&Fmt--4&clientld=l D

48

distortions introducedby depreciation.This claim was counteredby Stem Stewarton the

basis that depreciation is implicit in all IRR measures,CFROI included (Stewart, 1994, 71-84). pp. It is, however, important to understandthe limitations that CFROI has when usedto

38In the academicliterature performanceand wealth creation. measure and popularpress,

CFROI has been criticised for the following: First and foremost, its method of calculation, its rival measuressuch as EVA, is quite complicated hencemore difficult to explain unlike to managers. Second, as a form of an IRR, CFROI, by itself, does not provide any indication as to whether a firm is creating or destroying shareholdervalue. For instance,is CFROI of 11% good or bad? How much shareholder value has the firm created or a Without a benchmark or reference to a company's real cost of capital, it squandered? be impossible to answer these questions. Third, it requires one to make currentwould dollar adjustments, which results in a return on investment that is heavily sensitive to the these adjustments. Fourth, like EVA, in practice, there are many accounting quality of that can be made to estimate a firm's CFROI. However, the adjustmentsto adjustments EVA seem to make it more accurate, thus, more like a cash-basedmeasure; whereas the to CFROI appear to only make the measuremore understandable.Last adjustments made but not least, as CFROI is a non-linear measure,it may create a real communication issue For example, how much would a manager need to improve among non-financial people. flow to obtain, say, a 10% increasein CFROI? the cash

" For more details, seefor example,Young and O'Byrne (2001); Petersonand Peterson(1996); Clinton and Chen (1998); and Myers (1996).

49

2.3.6 The Balanced Scorecard (BSC)

A multi-dimensional framework for corporate performance known as "Balanced Scorecard", hereafter called BSC was developed and promoted by Kaplan and Norton (1992; 1993; 1996a; 1996b; 2001). It provides an integrated set of financial and nonfinancial measuresthat gives top managersa fast, but comprehensive view of a business. The financial measures include several accounting measures that report the results of that have already been taken (lagging indicators); whereas the non-financial actions measures encompass operational measures on customer satisfaction, internal Processes, and the organization's innovation, growth, and learning activities which will eventually lead to improved financial performance in the future period (leading indicators). By identifying and integrating both the financial and non-financial indicators of performance in a cause-and-effect relationship, the BSC becomesnot only a performance managementtool, but also a mechanism for translating strategy into action. Hence, the BSC has emerged as a strategic and control system that enablesbusinessunits to evaluate their operations from at least four different perspectives,by addressingthe following four (Kaplan and Norton, 1992, p.72): questions 1) How do we view shareholders?(Financial Perspective) 2) How do customers seeus? (Customer Perspective) 3) What must we excel in? (Internal BusinessPerspective)

50

4) Can we continue to improve and create value? (Innovation and Learning Perspective)

Given the above four questions, Hoque and James (2000, pp.2-3) suggest the following key measuresas indicators for the BSC usage. 9 Financial Perspective - includes profitability such as operating income, retum-oncapital employed, sales growth, generation of cash flow, or economic value added (EVA); * Customer Perspective encompassessuch measures as customer satisfaction,

customer retention, new customer acquisition, customer response time, market share,and customer profitability; e Internal-Business-Perspective - the key measuresinclude product design, product development, post-sales service, manufacturing efficiency, quality, etc.; and 9 Learning and growth perspective -measures the ability of employees, information system, and organisational proceduresto managethe businessand adapt to change. As Figure 4 shows that the balanced scorecardincludes a mix of outcome measures or lag indicators and perfonnance drivers or lead indicators. In practice, these indicators should be seen as a continuum. Kaplan and Norton (1996b, p.31) assumethe following learning and growth 4 causal relationships: measures of organizational measures of

internal business processes 4 measuresof customer perspective 4 financial measures. For instance, customer satisfaction is a leading indicator of EVA, but it may also be a lag

51

indicator of on-time delivery. In other words, better on-time delivery improvescustomer

in turn leads to higher EVA. On the other hand, while on-time delivery satisfaction, which is a lead indicator of customer satisfaction, it may also be a lag indicator of production time as well as the quality of both the manufacturing process and the products cycle themselves. Manufacturing process, product quality, rework rates, and cycle times are, in turn, lag indicators of employee skills and morale. Unlike any other accounting or valued-basedmeasure,the BSC does not provide an for company value. Its purpose is not to measurecompany value, but to aggregatenumber focus management on achieving the objectives that will result in value creation. Despite theoretical arguments and some empirical evidence favouring the adoption of BSC the implementation of this fairly sophisticated performance system is less worldwide, (e.g. Neely et al. 2004; Speckbacherand Pfeiffer 2003; Floque and James straightforward 2000; Anand et al. 2005; Ax and Bjornenak 2005). The BSC is essentially a conceptual have difficulties defining measuresas model, and as such, researchersand practitioners they are not clearly established (Ahn, 2001). It can be argued that Kaplan and Norton (1992,1996) have provided a simplistic picture of complicated world. Therefore, many have the feeling that, while the process of developing the balanced scorecardmay analysts be useful in identifying the value drivers of the company, the approach is difficult and Its complexity and subjectivity, therefore, make it awkward to measure and maintain. (Morin and Jarrell, 2001, p.335). unsuitable as a compensation criterion Although the BSC has gained a great deal of attention among corporate managers tool, little empirical evidence exists to substantiateclaims as a performance measurement

52

that it promotes superior financial performance compared to traditional performance (e.g. Davis and Albright 2004; Maltz et al. 2003). The critics of the measurementsystems BSC approach also argue that it lacks a long-term perspective; the distinction between

is blurred; and it is difficult to achievebalancebetweenthe financial and cause-and-effect

Norreklit (2000), for example, challenges the rationale of the non-financial measures. assumedcause-and-effect relationship arguing that a cause-and-effectrelationship should include a time lag between improving customers' satisfaction, learning and internal improvement on financial performance. Strack and Villis (2002) operations, and showing found that the BSC approach thrives in order to identify cause-and-effectrelationship but the linkages established are mostly qualitative. Furthermore, researchershave noted that the BSC does not contain a human resourcesperspective which, arguably, is a desirable dimension in any perfortnance measurement system (Maltz et al. (2003). strategic Nevertheless, the original appeal of the BSC approach for total business performance that it organized measurement under a small set of dimensions of measurement was businessperformance with which any manager could work, arguably (Kaplan and Norton, 1992).

2.4 The Association between EVA and Shareholder Value

In recent years, academics, corporate professionals, as well as the popular press have all shown great interest in the use of EVA as a measure of corporate successand A growing number of companies have adopted EVA or EVA-like measures value creation. key metrics for corporate performance and executive compensation. The as their

53

EVA have, consequently, come to realize it is the most reliable single-period proponentsof indicator of shareholder wealth. Stem Stewart, the promoters of EVA, for example, have

in the notion that it is superiorto traditional accounting-based asserted repeatedly measures

They have also proposed abandoning EPS, ROE and ROI which reflecting value creation. deem to be "misleading measures of corporate performance".39 They went even they is what drives stock prices".40 As a means of providing further to contend that "EVA for this claim, Stem Stewart conducted several in-house studies to justify such an support For example, Bennett Stewart III, a senior partner of Stem Stewart & Co., claims assertion. that: "EVA stands well out from the crowd as the single best measureof wealth creation basis [it] is almost 50% better than its closest accountingon a contemporaneous ... based competitor in explaining changes in shareholderwealth [and] as such, it ... be adopted with confidence as a company's primary internal financial can performance metric" (Stewart, 1994, p.75). However, surprisingly, to date, not enough empirical researchhas been done to support the (e.g., Ittner & Larcker, 1998, p.210). The relatively few empirical studiesthat above claims have addressed the issue, though, have provided some conflicting evidence about the EVA as a measureof corporate successand value creation (eg., Biddle et al., usefulnessof 1997; Chen and Dodd, 1997). Obviously, therefore, the superiority of EVA over commonly used performance is an empirical issue that ought to be explored and examined. In an influential metrics by Fortune magazine, Tully (1993) pointed out that EVA and stock prices article published
39

SeeStewart(1991), p.66. 1995,p.20. 40Stern Stewart advertisementin Harvard BusinessReview,November-December,

54

tendency of moving up and down together.To supporthis assertion, show a remarkable

Tully (1993, pp.43-44) quoted JamesMeenan, CFO of AT&T's long-distance firm, saying:

"We calculatedour EVA back to 1984and found an almost perfectcorrelationwith stock

" Not surprisingly, the very same author, five years later, in another article in Fortune price. his assertion that EVA is the best guide to stock prices and that it magazine, reiterated far better with stock performance than EPS does (Tully, 1998). By the same correlates token, Victor Rice, Chairman and CEO of Varity, a multi-billion dollar New York-based components and diesel engines, noticed that "At Varity, EVA manufacturer of automotive has become considerably more than just a yardstick. We fundamentally believe that, over time, there is a direct relationship between the improvement of EVA and a higher share So we have made EVA part of our mantra for building our corporate culture and price. for shareholders" (Rice, 1996, p. 40). creating wealth Of course, Stem Stewart has also carried out a number of empirical studies on the between EVA and MVA (e.g., Stewart, 1994; O'Byme, 1996; O'Byme, 1997; relationship Ross; 1997; Uyemura, Kantor, & Pettit, 1996; Ehrbar, 1998). Using the Performance 1000
4

Database,they found that popular accounting based earning measures,including bottomline net earnings after tax, earnings per share (EPS), and earnings growth rate, statistically 20 Percent of the changesin MVA recorded over the same period. It only 'explain' about found that the rate of return measuressuch as return on assets(ROA), return on equity also (ROE), and return on net assets(RONA), do have more explanatory power, and account for about 35 percent of the changes in MVA. However, according to Stem Stewart's

55

EVA scoredhigher than all of the other measures, explainingnearly 50 percentof analysis,
in MVA (see Table 2.1). the variation

In the words of Stem (1996, p.4), why the EVA correlation comes acrossas
is that "EVA, unlike ROE or RONA, takes into account the amount, as well as the stronger investment. [It] corrects for accounting distortions in GAAP income quality of corporate balance sheets,and specifies a minimum or required rate of return that must statementsand be earned on capital employed." The fact that EVA explains only 50% of the changesin MVA may not sound very impressive, but once we have ascertained that no other the change in MVA; it must be concededthat performance measurecan explain as much of EVA is definitely on the leading edge. The studies of Stem Stewart also reveal that each$1 increase in EVA, on average, brings a $9.50 increase in MVA (Ehrbar, 1998, p.78). Of Stem Stewart is hardly unbiased and the studies have yet to be vetted by money course, managersand researchers. In another research study, analyzing the computer industry from 1990-95, Milunovich and Tseui (1996) demonstrated the correlation between MVA and several frequently used financial metrics (including EVA). They found EVA to correlate other MVA somewhat better than other performance metrics. According to their analysis, with for 42% of the variation in MVA. EPS growth for 24%, ROE for EVA was able to account Cash Flow Growth for 25%, and Free Cash flow for 18%. Likewise, Uyernura, 29%, Free Kantor, and Pettit (1996) also studied the relationship between MVA and a variety of including EVA for the largest 100 bank holding companiesover the performance measures 1986 through 1995. They found that, among all the performance measures, ten-year period

56

EVA had the strongest correlation to MVA. According to their study, the statistical
between these performance measures and MVA are: EVA 40%, ROA 13%, correlation ROE 10%, Net Income 8%, and EPS 6%. Zafiris and Bayldon (1999) have further contributed to the emerging literature on EVA/MVA, by improving EVA's applicability as an advance corporate measure over The authors strongly believe that EVA has the potential to conventional accounting. becomethe standard single-period criterion measurefor decision making and performance However, for this to happen, the authors suggest a simple, but rigorous evaluation. approach to the EVA calculation based on up-to-date market values of the equity capital invested in the firm as opposed to book values. Unlike current practice in EVA measurement,the authors argue that this approach opens up "new ways of estimating relevant true opportunity costs generally making greater use of benchmarks derived from the competitive environment of the particular firm, as opposed to the firm
itself.,, 41

Moreover, the authors drawing from the debates and concepts already established within the framework of neoclassical economics, have successfully put forward in the paper a the EVA criterion suitable for both decision making purposes (ex ante) and for version of (ex post). measurementof achievement Using a sample of 452 firms during the period 1985-1994, Lehn and Makhjia (1997) examined the relationship between EVA/MVA and stock returns. They found that highly correlated with averageEVA than with other averagetraditional stock returns more (ROA), return on equity (ROE), or return performance measures such as return on assets
41SeeZafiris and Bayldon (1999), p.95.

57

(ROS). In addition, they found EVA performedsomewhat betterthan accountingon sales in predicting CEO turnover. Likewise, Worthingtonand West (2004) useda based metrics
I 10 Australian companies over the period 1992-1998to examine whether EVA sample of

is more highly associatedwith stock returns than other traditional accounted-based

The authors found that stock returns are more closely correlated with EVA than measures. income, earnings, and net cash flow, respectively. residual In another attempt to assessthe strength of the MVA and EVA relationship, Grant (2003) calculated regression statistics between the MVA-to-Capital ratio (dependent the EVA-to-Capital ratio (explanatory variable) for the 50 largest U. S. wealth variable) and 2000. As expected, he found that a linear relationship creator corporations at year-end between these twin measures of corporate performance and value creation. His exists demonstrated that when the EVA-to-Capital ratio is large and positive; the evidence MVA-to-Capital ratio will also be high and positive. Likewise, when the corresponding EVA-to-Capital ratio is low or negative, the corresponding MVA-to-Capital ratio is also low. These cross-sectional regression statistics reveal a statistically significant relationship between these two measures of corporate financial success.With an EVA "beta" (slope 35.93, and a t-statistic of 3.53, the EVA-Capital ratio for large U. S. wealth coefficient) of is a highly significant variable in the MVA equation (Grant, 2003, p. 86). creators
42

The

19% of the movement in the MVA-to-Capital ratio among topstudy also reveals that

42The predicted MVA-to-capital ratios for theselarge capitalizationfirms were estimated: MVA/Capital = 3.36 + 35.93 EVA/Capital (3.53) (3.49) (t-value) Adjusted RA2=19% N=50 Firms

58

be explainedby contemporaneous variationsin the S. firms at yearend-2000can U. ranked

EVA-to-Capital factor.

to reassess value relevanceof EVA with respect other the to In yet anotherattempt Feltham et al. (2004) in a more recentstudy upheldthe useof EVA. metrics, performance
They tried to examine whether the results in Biddle et al. (1997) which suggestedthat EVA still hold true. To preserve the integrity of their analysis,authors earningsoutperform the exact statistical and econometrics procedures as Biddle et al., except with replicated

different sets of companies,different time periods, and different markets.The resultsof

their replications, in general, were not consistent with the findings of Biddle et al. Thus, they disassociatedthemselves from Biddle et al's assertion and concluded that EVA does in fact beat earnings. However, Feltham et al. did suggest that the debate should be reopened. Further to this, certain studies have suggestedanother tack -- that EVA is predictive However, it is not the only performance measure tied directly to of stock returns. by its proponents (e.g., Stewart 1991). Yet again, in still shareholder wealth as claimed have even found an adverse relationship between EVA and other cases, some studies Using a sample of 566 companies from 1983-1992, Dodd and Chen shareholder wealth. (1996), for instance, found that return on assets(ROA) explained stock returns better than EVA with R2 of 24.5%. The R2 for other metrics showed that EVA accounted for 20.2%; Residual Income for 19.4%; and ROA, EPS, and ROE for approximately 5-7%. Their empirical results demonstratedthat stock returns and EVA per approximately by EVA advocates; however, the association was far from as touted share are correlated

59

Clinton and Chen (1998) further comparedEVA's ability to explain stock returns perfect. host of other "traditional reported, residual-based,adjusted, and cash-based" with a Their findings also proved that EVA is the only measure that does not measures.
consistently reflect stock returns. Biddle, Bowen, and Wallace (1997) found similar results. Their findings are in support of a simple earnings measure. For example, they found that overwhelmingly highly associated with market-adjusted returns (R2= earnings are significantly more income (R2=7.3%), EVA (R2=6.5%) or operating cash flows 12.8%) than with residual 2=2.8%). (R In their article, Biddle et al. performed a fairly comprehensive test on the

value relevance of all the components of Stem Stewart's EVA. They broke down EVA into its component parts as follows:

EVA = CFO + Accruals +A TInt - CapChg + AcctAdj

(2.18)

where,
CFO Accruals ATInt CapChg AcctAdJ flow from operations = cash depreciation = accounting accruals, such as the after tax interest expenses = the capital charge of all invested capital = Stem Stewart's capital adjustments(i. e., assetre-valuation) and = the adjustment of operating profits.

60

The authors examined the relative and incremental information content of each
EVA in an attempt to find out whether EVA, or any of its components,is component of

highly associatedwith stock returns and firm value than accounting earnings, more
income, or cash flow from operations. They found that accounting earnings, in residual have the highest association with stock returns and firm value than EVA, residual general, income, or cash flow from operations. Moreover, they found that EVA componentsonly insignificant incremental value beyond accounting earnings in explaining stock add Consequently, they conclude that their findings reject the claim that EVA returns. dominatesearnings and suggestrather that earnings generally outperform EVA. Chen and Dodd (1997), likewise, examined the explanatory power of various EPS, ROA, ROE, RI, as well as various EVA related accounting measures such as Using a sample of 605 companies taken from the 1992 Stem Stewart 1000 measures. during the period spanning 1983-92, Chen and Dodd found that "EVA is a useful database However, EVA is neither as perfect as claimed by its measureof corporate performance. is it the only performance measurethat suggestsa path to a superior stock advocates,nor " (p.319) retum.

2.5 Concluding Remarks

A review of the pertinent literature regarding performance measures and value is presented in this chapter. Particular attention is paid to the six most widely used creation discussed value-based performance metrics in the business world and academic and be concluded that the existing bulk of empirical evidence indicates that literature. It may

61

EVA's

superiority

over

commonly

used accounting and non-accounting-based

has not yet been fully established. Hence, further empirical studies performance measures

this highly complicatedissuein moredepth. to arewarranted explore

62

CHAPTER 3 RESEARCH ISSUES AND HYPOTHESES

3.1 Introduction
The purpose of this study is to empirically test the hypothesis whether there is a statistical association between EVA and shareholder wealth. Economic Value significant Added (EVA), a surrogate for abnormal profit in the economist's sense, has received a deal of attention as another new single-period criterion for decision making and great performance evaluation (e.g., Zafiris and Bayldon, 1999). It has even been predicted that it will replace earnings per share (EPS) as the most valuable financial indicator in the Wall Street Journal's regular stock and earnings reports (Zarowin, 1995). Moreover, it has been repeatedly portrayed by Stem Stewart and other proponents as the key to creating shareholderwealth (e.g. Tully 1993; Ehrbar 1999; Grant 2003; Stewart 1991). Hence, it is important to empirically investigate how much EVA can explain MVA and to find out whether it is a reliable guide for achieving the goal of being able to maximize shareholder wealth. Two commonly-used performance measures,namely TSR and Tobin's Q are also in order to highlight the value-relevanceof EVA vis-a-vis these measures. considered The remainder of the chapter is organized as follows. Section 2 presents the the link between shareholder value (SHV) and various empirical questions concerning including EVA. Section 3 develops and identifies the main performance metrics, hypothesis of the study and the associated arguments about the link between EVA and SHV, and Section 4 concludes.

63

3.2 Research Issues

Academic literature indicatesthat links betweenvariousperformance measures and
(SHV) are probable; however, this can only be definitively concluded by shareholdervalue further empirical tests on these relationships. In order to determine whether performing

there is any associationbetweenthesemeasures and the market value of a firm, a direct

link between alternative performance measures and the value created for shareholders first be formulated. In this study, the firm's market value added or MVA would should the dependent variable in thevarious regression models and the regressions' Rserve as squares(R) are then used as a gauge of the information's usefulness to the independent

variables.
Several empirical questions concerning the link between shareholder value (SHV) including EVA will be developed and examined. This and various performance metrics, is primarily intended to test whether a new value measurementparadigm such as study EVA better explains the variation in market value added (MVA) compared to traditional total shareholder return (TSR) and Tobin's value-basedperformance measuressuch as Thus, this study's analysis consists of three closely related empirical questions. 1) Does a statistical relationship between EVA and shareholder wealth exist, if it does, how much of the variation of the shareholder value can be and by EVA? explained 2) Does EVA dominate traditional value-based performance measures such as TSR

MVA? Tobin's Q in explainingcontemporaneous and

64

3)

Do components unique to EVA, such as net operating profit after tax (NOPAT), return on invested capital (ROIC), cost of capital (WACC), and total invested (TIC) help in explaining contemporaneous MVA beyond the explanation capital by traditional performance measures? given

3.3 Research Hypotheses

Generally, in a regression model, the null hypothesis is set up as (HO:Pi = 0). In our case, this means that the associated explanatory variable, that is, EVA has no effect on MVA (the dependent variable) and that knowledge of EVA would not help in explaining the variation in MVA. The alternative hypothesis is set up as (HI: Pi:t 0) which statesthat the slope of the regression line is not equal to zero; that is, EVA and MVA are linearly related. This implies that knowledge of the value of EVA does provide information conceming the associatedvalue of MVA. To address the previously mentioned research questions, this study examines the following hypotheses stated in the (alternative form).

3.3.1 Hypothesis One:

Hl:

Economic Value Added is significantly firm's Market Value Added.

and positively associated with the

This hypothesis directly addressesthe first study's question and examines whether is a significant relationship between Economic Value Added and changes in there

65

The coefficient on EVA metric is viewed as the weight that stock wealth. shareholder
to this measure. A positive sign of the coefficient indicates that EVA is market attaches

the dependent variable. Consistent with prior empirical studies that associatedwith
performance measures(e.g., Grant 1996,2003; Kramer and Pushner evaluate alternative 1997; Lehn and Makhija 1997), this study uses the association with the firm's Market

Value Added (or MVA) asthe criterion to determinethe bestmeasure.

This study hypothesizes that EVA is strongly and positively correlated with MVA it provides additional information to explain the variation in the Market Value and as such Added of the firm. This study's prediction is consistent with the theoretical valuation in finance and accounting which suggest that various components of residual models income should be associated with firm value in a manner that differs predictably in terms both sign and magnitude of the association, and that they depend on the accounting and of in which a firm operates(e.g., Livnat and Zarowin 1990; Barth et economic environments 1999). al.

3.3.2 Hypothesis Two:

"EVA dominates value-based performance measures such as TSR and Tobin's Q in MVA. " explaining contemporaneous

H2:

EVA, TSR, and Tobin's Q in This study compares the value-relevance of MVA. It is hypothesized that the three value-basedmetrics are positively and predicting MVA and as such could serve as important predictors of MVA. highly correlated with EVA would outperform TSR and Tobin's Q in Moreover, the study is also predicting that

66

the variation in the market value addedof the firm. This hypothesisis in line explaining finding that EVA is more highly associated with recentstudies with stock returnsand firm
than accrual earnings, residual income, or cash flow from operations (e.g., Feltham values 2004; Lnhn and Makhija 1997; Worthington and West 2004). et al.

3.3.3 Hypothesis Three:

H3:

Components unique to EVA help in explaining contemporaneous MVA beyond the explanation given by value-based performance measures such as TSR and Tobin's This hypothesis assesseswhether components unique to EVA model such as

NOPAT, ROIC, WACC, and TIC have greater value-relevance (or infonnation usefulness) over TSR and Tobin's Q in explaining the variation of the firm's market value added.The rational for decomposing EVA is to examine whether "aggregate" EVA masks much of the its individual components, and whether the disaggregation of EVA improves usefulnessof the degree of association with MVA. Prior works suggestthat EVA components add only to information content beyond traditional accounting-based measures (e.g., marginally Biddle et al. 1997). This study predicts that EVA components are expected to have incremental value-relevant information beyond that contained in TSR and Tobin's

67

3.4 Concluding Remarks

The preceding chapter (chapter 2) focused on reviewing the theories and the findings of other studies in the literature on performance measuresand empirical
This chapter (3) has identified three hypothesesabout the link shareholdervalue creation. between EVA and SHV. These hypotheses and their associated arguments were also

highlighted and discussed. This study hypothesizes that EVA is strongly and positively

with MVA and as such it provides additional infonnation to explain the correlated
in the Market Value Added of the firm. The study also predicts that EVA would variation TSR and Tobin's Q in being able to explain the variation in the market value outperform the firm. Finally, this study shows that EVA components are expected to have added of incremental value-relevant information beyond that contained in TSR and Tobin's

68

APTER 4 PJESEARCH DESIGN AND METHODOLOGY

4.1 Introduction
The previous chapter identified and discussed three empirical hypotheses about EVA and shareholder value creation. This chapter outlines and describes the research design and methodology used in testing the hypotheses.As well, it introduces the primary defines its variables. It also delves into the data sources and the specific model and that link different performance measuresto the dependent variable of regression models MVA. This chapter is structured as follows: Section 2 delineatesthe data sourcesand the techniques used in this study. In Sections 3 and 4, the primary models to test the statistical hypothesis are presented and their variables defined and explained in terms of empirical their theoretical relevance. Concluding remarks are provided at the end of the chapter.

4.2 Data Sources and Statistical Techniques

Two separatesourcesof data have been used: 1) The 2002 US EVA/MVA Annual 1000 Ranking Databasecompiled by New York-based financial consulting firm Stem Stewart & Co. the 2) The DATASTREAM files. Performance 1000 is a file that includes 20 years of historical annual The Stem Stewart NOPAT, Capital, Return on Capital, Cost of Capital, Market Value, data for MVA, EVA,

69

CompanyType, Industry Index, and Three-Five- and Ten-year ShareholderReturnsfor

1000 top publicly owned US industrial and non-financial service firms (see Appendix

The data is published on an annual basis and can be obtainedfrom Stem Stewart& Co.,
New York
43

Data for MVA along with EVA and its components over the period of 1991-

2002 was taken directly from the Stem Stewart Excel file or calculated using data provided

by this source.Furthermore,to obtain the datausedto estimateTobin's Q valuesandTSR,

the DATASTREAM files were examined for those fiscal years. This study uses panel data (or sometimes referred as pooled data) to test the hypotheses. It can be considered an unbalanced panel data, since not all firms research data for all years between 1991 and 2002. The panel data analysis is an advanced provide analytical technique that captures not only the variations of a single firm over time and variations of many firms at a given point in time, but the variations of these two dimensions simultaneously (e.g., Hsiao 2003; Baltagi 2001; Pindyck and Rubinfeld 1998). This simply means that panel data analysis combines both a cross-sectionaldata (N), and a time-series data (T) to produce a data set of (N x T) observations; where N can be a 44 firms, households, countries, etc. specific set of In the last decade or so, panel data analysis has became central in quantitative Its popularity has been greatly increased among social and behavioural science studies.

it becameone of the most active and innovative bodies of literature in researchers and

43hqp://www. sternstewrt. com dominated;that is, whereN is significantly "Panel data usually refers to data which are cross-sectionally larger than T. Such data usually have a fixed T, andthe asymptoticsare in N.

70

45. For economists econometrics. and social scientists, panel data sets provide several distinct advantages over conventional cross-sectional or time-series data sets. Baltagi (200 1, pp.5-7), for instance, lists the following advantagesfrom using panel data: 1) 2) for individual heterogeneity; controlling data provide a more informative data set, more variability, less panel collinearity among the variables, more degrees of freedom and more efficiency; 3) 4) data are better able to study the dynamics of adjustment; panel panel data are better able to identify and measure effects that are simply not detectable in pure cross-sectionor pure time-series data; panel data models allow us to construct and test more complicated behavioural models than purely cross-sectionor time-series data; and 6) panel data are usually gathered on micro units, like individuals, firms and households. The data for this study are annual observations,collected over a period of 12 years 1,000 firms, from 1991-2002. A pooled regression of observations from all 1000 firms on time period would yield estimatesbasedon both firms and time periods. With N crossand firms observed for T time periods the total number of observations in the data set sectional be NxT= will 1000 x 12 = 12000 panel observations. Thus, the pooled data structure

firm-year as unit of analysis yield a substantially larger number of observationsthan using is possible with either individual firm time-series or cross-sectionalanalysis.
45 See,for example(Greene,2008; Dougherty,2007).

71

4.3 Model Specification

It is worth noting that the purposeof this study is not to build an elaborate model for shareholdervalue, but only to examine and highlight the value-relevance of several
performance metrics vis-a-vis EVA in predicting shareholder wealth. Hence, a

but only a combination of univariate46 sophisticated econometric analysis was not used, and multivariate regression analysis to perform the required
47 teStS.

The statistical models

in this study were an eclectic combination of models used by Biddle et al. (1997), used Chen and Dodd (1997), and Kramer and Pushner (1997). The data for this analysis were two dimensional with respect to time and cross section. The model used in this study, therefore, had to include the same dimensions in order to produce a pooled design. All regressionswere computed using the 'Regression" routine in STATA, version 9.1. In this study, two regression models were suggested to test the empirical hypotheses. The first model examines the association between alternative corporate performance metrics and the MVA as well as highlights the value-relevance of these in explaining firm value and shareholderwealth (Hypothesis one and competing measures two).

46 univariate model has only one right-hand-sidevariable. A 47It is important to note that multivariate models allow us to estimaterelationswhere two or more dependentvariable. On the other hand,the interpretationof the parameters independent of variables affect a basedon the ceterisparibus assumption-that one of the independent any multivariate regressionare is changed,with all other held constant,to producethe measured variable. effect on the dependent variables For further information, seefor example,Schmidt (2005).

72

IMVAjl + a, EVA,, + ell = ao MVAj,

=go

+ATSR, + u,

MVAj, = yo + yj Tobin's Q,, + v,, MVAj, = iro + ir, EVA,,+

; T2

TSR,,+/ 'r3Tobin'sQ,, + V/,,

The dependent variable here is the market value added (MVA) for firm i in period t. The in this model are: economic value added (EVA), total shareholder explanatory variables return (TSR), and Tobin's Q. Positive and significant coefficients are hypothesized. All in the above model are scaled by the beginning-of-period total invested capital variables (TICt-1)to mitigate heteroscedasticity. The second model investigates whether EVA components can explain

MVA beyond that explained by others performance metrics (Hypothesis contemporaneous

three).
Af VAil = 80 +Aprofitability spread,, + e,,

AIVA, 80 +AL agged EVA,, + e, I=

IMVAj,=A+A

total invetsed capital,,+ e,,

IAlfVA,,= 80 + 81 TSR,

IMVA,, =, 8,, +A Tobin'sQ,,+ e,,

JMVAjf=, 60 +Aprofitability +, 82Lagged EVA,, +, 83total invested capital,, spread,,

+A +,84TSR,, Tobin'sQj,+ e,,

in In this model, EVA is broken down into three components order to evaluatethe
toward explaining contemporaneousMVA. The dependentvariable is contribution of each

73

MVA. The explanatory variables are: profitability spread (or EVA 48 given as spread),

laggedEVA (LagEVA), total investedcapital (TIC), TSR, and Tobin's Q. All variables are
defined and are also scaled (normalized) by lagged total invested as previously capital (TICt-1) to mitigate heteroscedasticity. To account for any non-normality in the variables, the values are expressed in natural logarithms. This model is also estimated using the least squaresregression and positive coefficients are hypothesized. pooled ordinary

4.4 Research Variables

In addition to the dependent variable, Market Value Added (MVA), there are three independent variables: Economic Value Added (EVA), Tobin's Q, and Total categoriesof ShareholderReturn (TSR). The following subsectionsdescribe all of these variables.

4.4.1 Dependent Variable: The dependent variable chosen for this study is Market Value Added (MVA) -- a commonly used variable to measure corporate performance and value creation. As MVA is a concept developed by Stem Stewart & Co. and may be previously mentioned, defined as the aggregate net present value (NPV) of all the firm's activities and investments. It represents the value created (or destroyed) over the lifetime of a firm and be seen as a proxy for the past and current value of the firm's strategies. MVA is can

48Profitability spread(SPREAD) refers to the difference betweenthe return on investedcapital (ROIC), and (WACC), where ROIC is calculated by dividing NOPAT by total the weighted average cost of capital invested capital in the company at the beginning of the year. In literature, the spread between ROIC and WACC is often referred to as (1) the "residual return on capital," (2) the "surplus return on capital," (3) the "excessoperating return on investedcapital," as well as, (4) the "EVA spread". SeeGrant 2003, p.23.

74

the difference betweentotal market value and total "economic" book calculatedas value (capital). Total market value is defined as an approximationof the fair marketvalue of a firm's entire debt and equity capitalization; whereas,total "economic" book value is
defined as a firm's net assets with some adjustments (Stewart 1991; Stem 1996). This is in the following equation: MVA, MV, BV,; expressed where MVA = market value = added, MV = total market value of the firm, BV = total 44economic value", and the t subdenoting the time-period. script MVA was chosen simply becauseit is a measureof value creation that reflects the cumulative wealth created and as such is expected to reflect both tangible and intangible value. It is superior to simple market value as a performance measurebecauseit removes the capital employed in the firm away from the cumulative value created to demonstrate how well management has utilized its resources. One important methodology issue, though, is whether the level of MVA is influenced by the size of a firm. In order to control for a firm's size, MVA is scaled by dividing it by the total capital invested in the company at the beginning of the year (TICt-1) -- a variable that is available from the Stem Stewart
1000

database.

4.4.2 Independent Variables The independent variables used in our analysis are: Economic Value Added (EVA); Total Shareholder Return (TSR); Tobin's Q or Market to Book Value (MBV); Net Operating Profit After Tax (NOPAT), Return on Invested Capital (ROIC), Weighted Average Cost of Capital (WACC), Total invested Capital (TIC), and Profitability Spread

75

(or EVA Spread). All these measuresare frequently used in practice as corporate and performance value creationmetrics.
Economic Value Added: EVA is a measure of true 'economic' profit, or the

by which earningsexceedor fall short of the requiredrate of return that investors amount
to earn when investing in other assetsof comparable risk. A senior partner and can expect

in Stem Stewart& Co., G. BennettStewart111, his 1991book 'The Questfor in co-founder

Value', defined EVA as the difference between a firtn's net operating profit after tax (NOPAT) and an appropriate charge for the opportunity cost of all capital invested in that firm (Stewart, 1991). Since traditional earnings-basedmeasuressuch as ROA understate the cost of capital by ignoring the opportunity cost of equity capital, EVA is designed to take this into account. EVA is calculated by multiplying the amount of invested capital (TIC) by the spread between the return on invested capital (ROIC) and the cost of capital (WACC), which is the required or minimum rate of return a firm must earn to compensate its investors for the risk they bear (Stem, 1996). This is expressed in the following EVA, = NOPAT, - (WACC, * TIC, ) = [ROIC - WACC], * TIC, where NOPAT equation: -, -,; tax, WACC = weighted averagecost of capital, ROIC = return = net operating profit after invested capital = (NOPAT/ TICt-1), TIC = total invested capital, and the t sub-script on denoting the time-period. To properly estimate EVA, the total 'book capital' invested in a company as well as income has to be adjusted. Table (4.1) shows Stem Stewart's the after-tax operating "bottom up" and "top down" approachesto estimating a firm's unlevered NOPAT while list of potential adjustments in the equivalent "asset" and "financing" Table (4.2) shows a

76

firm's investedcapital 49Theseadjustments necessary to to approaches estimatingthe are . potential anomaliesthat might be introduced by conventionalaccounting eliminate any intendedto producean adjustedbalancesheetthat reflects the economic practicesand are
than the inherently conservative, historical-costvalue of assets-in-place more accurately basedbalance sheet guided by GAAP. Like the MVA case, the information about EVA is from the Stem Stewart 1000 Database. also obtained

Tobin's Q or Market-to-Book Value (MBV): Tobin's Q is a measureof the real value

by management and is defined as the ratio of the market value of a firm to the created its assets. However, because of the unavailability of data, the replacement cost of

the "Q" value has been renderedtoo complex and cumbersome(e.g., calculation of
Damodaran, 2002). Therefore, several proxies for Tobin's Q have been suggestedin the literature (e.g., Chung & Pruitt 1994; Perfect and Wiles 1994). One approximation that to be empirically close to the "Q" value obtained via the Lindendberg and Ross' seems

model5o

is

proxy

given

by

Chung

and

Pruitt

(1994):

ApproximateQ, = (MVEI + PSI + DEBT,)/ BTA,; where, MVE is the product of a firm's
PS is the liquidating share price and the number of common stock shares outstanding; firm's outstanding preferred stock; DEBT is the value of the firm's short-term value of the its short-term assets, plus the book value of the firm's long-term debt; liabilities net of

49For a more comprehensive discussion of EVA-based accounting adjustments, see Stewart (1991). 50See Lindenberg and Ross 0 98 1).

77

BTA is the book value of the total assets a firm; and the t sub-scriptdenotingthe timeof period. In this approximation,Tobin's Q is the ratio of the market value of a firm to total as such differs from the Lindendbergand Rossmodel in that approximate and assets
implicitly assumes that the replacement values of a firm's plant, equipment, and

inventories are equal to their book values. Clearly, the simplified procedures involved in

the calculation of approximate"Q" representa compromisebetweenanalytical precision

and computation (Chung and Pruitt 1994). On the other hand, all the required inputs to calculate this ratio are readily obtainable from a firm's basic financial and accounting information, so the DATASTREAM file has been used as the data source for the

calculation eachfirm's Q ratio. of

Total Shareholder Return: The TSR metric is the most direct measure of changes in in percentage terms. It can be easily shareholder value over a given period, expressed compared from firm to firm, and benchmarked against industry or market returns, without having to worry about size bias (Young and O'Byrne 2001). Shareholderscan earn a return their investment in two ways: through dividends and through stock price appreciationor on depreciation. Over a given period of time, t, the total shareholderreturn (TSR) for firm be specified as:TSRj, = [DPSjl+l + PPSjl+l - PPSj, ]/ PPSj,; where DPSjt+ dividends can I= the end of period t, PPSjt= price per share at the beginning of period t, per share paid at PPSjt+l= price per share at the end of period t. The DATASTREAM file has been and the data source for the calculation of TSR. used as

78

4.5 Concluding Remarks

The purposeof this chapteris to describe research the methodologyandthe sources datathat were used in this study. The first sectionhas beendedicatedto describingthe of databases, sample,as well as the rationale behind the statistical techniques the that were
implemented in this study. The second section of this chapter specifies the empirical model to investigate the usefulness of EVA as a measure of performance and value employed The third section was devoted to describing the measurementof variables. The creation. following chapter provides descriptive statistics on select variables and reports on and discussesthe empirical results of testing the researchhypotheses.

79

CHAPTER 5 DATA ANALYSIS AND PRESENTATION RESULTS

5.1 Introduction
While chapter 4 introduced the statistical models used in this study on EVA and defined their variables, this chapter further studies the relationship between EVA SHV and by testing the proposed statistical models using regression analysis. The empirical analysis of this thesis focuses on the contemporaneous relationship between MVA and three commonly used performance metrics with the primary focus being on EVA. Specifically, 51 this thesis investigated the value-relevance of these competing performance metrics in explaining SHV. Therefore, the thesis has provided empirical evidence whether the current period realization of EVA is more closely associatedwith SHV than are other traditional performance measuressuch as TSR and Tobin's This chapter then provides descriptive statistics on selected variables as well as discusses the empirical results that stem from the testing of the research reports on and hypotheses.Starting with descriptive statistics, the distribution of the data in terms of the the standard deviations has been examined closely. Further to this, the idea means and by EVA's proponents that it is a superior measure of shareholder value creation asserted

OF STATISTICAL

" Also called "information content", seefor exampleBiddle et al. (1995).

80

also tested empirically. Finally, concluding remarks are provided at the end of the was

chapter.

5.2 Descriptive Statistics

As previously mentioned, the empirical tests in this study rely on a pooled desig., the data were analysed by the econometric software named STATA (Version 9.1). The and data are annual observations, collected over a period of 12 years of 1,000 firms, from 1991-2002. The dependent variable for this study is MVA and the explanatory variables are a set of performance metrics including EVA. These variables calculated on a twelfthbasis resulted in a final usable pooled data sample of 8945 year-firm year average

observations.
Table 5.1 presents the means and the standarddeviations for the primary variables in this study. From the Table, it should be noted that while MVA and NOPAT examined (about $5266.249 million and $401.9223 million respectively), the are positive on average level of EVA is negative (-$81.15118 million), which demonstratesthe significance mean the cost of capital (WACC) and implies significant growth expectationsfor future EVA. of On the other hand, Table 5.2 illustrates the relationship betweenMVA and the independent The correlation coefficients thus reveal a significant association between MVA variables. EVA variables, suggesting that EVA and its components separately yield and all of the information that is perceived important by the stock market, a rightful claim by EVA Nevertheless, the relationship between MVA and the EVA measureis far from advocates. 0.6102 between MVA and EVA indicates that increasing EVA perfect. A correlation of

81

is not all that matters in the marketplace since only 37-23% of the variation in MVA alone

be explainedby the measure. can

52

5.3 Test of Hypothesis

The objective of this study is to test the hypothesis that EVA is highly associated MVA. It is not the purpose here, though, to explain the determinants of MVA, but with to show how well EVA acts as a genuine explanatory variable for MVA, in order to only justify its appropriateness for performance measurement, CEO compensation, financial and shareholder value creation. As such, sophisticated econometric models or reporting, have not been employed, but instead a combination of univariate and adjustments to compare EVA with other single-period multivariate regression analysis was used performance measures. 5.3.1 Hypothesis One "Economic Value Added is significantly and positively associated with the firm's Market Value Added. " Hypothesis I was tested using the following univariate regression model with the dependentvariable of MVA scaled by beginning-of-year invested capital (MVAjt/TICjt-i)

than the determination(R2) is a more meaningful measure 52 the regressioncontext, the coefficient of In variable The former tells us the proportion of variation in the dependent (R). coefficient of correlation (s) and thereforeprovidesan overall measure the extentto which the of by the explanatory variable explained in the other. The latter doesnot havesuchvalue. Also note in one variable determinesthe variation variation (R). Seefor example,Gujarati (2003). that R2can be computed as the squaredcoefficient of correlation

82

the independent variable of EVA scaled by beginning-of-year invested and capital (EVA,-t/TlCit-,).

, + 2 sEVAi, +ei, sMVAil =

(5.1\

where,
N1VAjt= market value added for firm i in period t sMVAjt =standardized MVA=(MVAit / TICit-1) EVAit= economic value added for firm i in period sEVAit=standardized EVA=(EVAit/ TICit-1) TICit-1--total invested capital for firm i in period t- I [N(O, is a random disturbance term eit= c'j This model was estimated using a pooled least square regression. The overall model yielded a positive and statistically significant coefficient of 10.33 and an R2 of 0.3724 for the entire sample (see Table 5.3, Panel A). This upholds Hypothesis I that EVA is positively and significantly related to MVA. The very low p-value (0.000) implies that the EVA coefficient is statistically significant that allows for the null hypothesis to -a result be rejected in favour of the alternative hypothesis. Moreover, the positive sign on the EVA high t-statistics of 75.34 indicates that EVA has a strong coefficient along with sufficiently MVA. For each dollar increasein EVA, there would be $10.33 increasein MVA. effect on This should come as no surprise as a number of past empirical tests of the relationship between EVA and MVA show similar results. For example, in one of Stem Stewart's

83

using the Performance 1000 Database, EVA was found to statistically 'explain' studies 50% of the changes in a company's MVA. The research also showed that each $1 about increasein EVA brings, on average,a $9.50 increasein MVA (Ehrbar, 1998, p.78). While these results are significant, much of the variation of MVA remains In order to obtain more insight into the strength of EVA as a proxy for MVA, unexplained. least squares regression was performed with net operating profits after taxes an ordinary (NOPAT) as an independent variable:

AsNOPA Ti, + e,, sMVA,, =, + go where, MVAit= market value added for firm i in period t MVA=(MVAit / TICit-1) sMVAjt =standardized NOPATit= net operating profit after tax for firm i in period t NOPAT=(NOPATit / TIC it.,) sNOPATit =standardized TICit-1--total invested capital for firm i in period t-I [N(O, is a random disturbance term a'] eit=

(5.2)

5.3, Panel B. Compared to Table 5.3, Panel A, it can The results are summarized in Table is positively related to both EVA and NOPAT in the same be seen that the level of MVA the total variation in MVA than EVA However, NOPAT explains slightly more of periods.

84

53This suggests does. that the level of NOPAT is not only a better proxy, but it is also a
better predictor of corporate performance than the level of EVA. 5.3.2 Hypothesis Two: "EVA dominates value-based performance measures such as TSR and Tobin',

in explaining contemporaneousMVA. 11
To test for the incremental value-relevance (also called information usefulnessor content) of economic value added (EVA) over the value relevance of total shareholder (TSR) and Tobin's Q, the following multivariate regressionmodel was used: return

In MVA,, = flo +Aln

E VAj, +, 62 In TSR,,+, 63 In Tobin's Q,, + e,,

(5.3)

This model was also estimated using a pooled least square regression. The dependent variable was MVA for firm (i) in period (t); whereas,the explanatory variables were EVA, TSR, and Tobin's Q. To account for any non-normality in the variables, the values are in logarithms. Both the dependent and the independent variables were expressed by the beginning-of-year invested capital to reduce heteroskedasticity. normalized Table 5.4, Panel A, shows the estimated coefficients, standarderrors, t-statistics, FR-square for this model (Eq.5.3) and indicates that all three performance statistic, and is, EVA, TSR, and Tobin's Q are positively associated with changes in metrics, that (MVA). The coefficients for EVA, TSR, and Tobin's Q are 0.2410, shareholder value
" Since the only difference between EVA and NOPAT is the cost of capital (COC), results favouring NOPAT may be attributable to mis-estimation by Stern Stewart of the cost of capital (potentially from using See for example, Kramer and Pushner (1997) and Zafiris a CAPM approach to estimate the cost of equity). and Bayldon (1999).

85

0.07629, and 0.2897 respectively and all are significant at 5%. To assessor check for Belsley (1991) and Belsley et al. (1980) recommend looking at 'the multicollinearity,

is the squareroot of the ratio of the largestto smallest condition number' or y, which root: characteristic
1/2

max root min root

(5.4)

Belsley et al. suggest that condition numbers larger than 30 indicate serious 54 Some numbers larger than 100 imply severe multicollinearity. multicollinearity, while indicator of multicol linearity.55 the Variance Inflation Factor (VIF) as an researchersuse The VIF is a measure of the strength of the relationship between each explanatory variable in the regressionand can be defined as: and all other explanatory variables VIFJ = IR (I
2)

(5.5)

It shows how the variance of an estimator is inflated by the presenceof multicollinearity. If (ideal case), then R, =0 and VIF will be 1. As a rule of thumb, if there is no collinearity 10, a severemulticol linearity problem may be the VIF of an independent variable exceeds inverse of the VIF, which is called Tolerance (TOL). indicated. Finally, one can use the That is, ToLi =I= VIFJ (I - Ri) (5.6)

54 further details, seeGreene(1993), p.269. For " See,for example,Dielman (2005), pp. 161-164.

86

Thus, when Rj =I

(i. e., perfect collinearity), TOLj =0

and when Rj =0 (i. e., no

TOLj = 1. In other words, the closer TOLj is to collinearity whatsoever), zero, the greater the degree of collinearity of that variable with other regressors.Becauseof this intimate between VIF and TOL, researchers, connection quite often, use them interchangeably. Examination of the VIF reveals that there is no serious multicol linearity in problem the regression model. As shown in Table 5.4, Panel B, the highest VIF in this model is (1.28). This suggeststhat, while multicol linearity, exists, it does not pose a problem. Again in Table 5.4, the results of this regression not only provided strong as summarized the significance and direction of the relationships as previously hypothesized; evidence of but also established a baseline by which to analyze the incremental value-relevance of EVA. As previously mentioned, the data under study is panel data sometimesreferred as data and consists of a combination of time-series and cross-sectionaldata. The two pooled frequently estimation techniques used are the Fixed Effects Model (FE) and the most Random Effects Model (RE) (see, Wooldridge, 2002). To choose between the fixed or the Hausman (1978) specification test was used. The test statistic random-effects models, is asymptotically distributed as Chi-Squared and the test is based on the Wald criterion (Greene, 1993, p.480). The null hypothesis underlying the Hausman test is that the fixed the alternative, the fixed effect and random specifications are consistent, whereas under is not. As can be seen from Table 5.5, the test is, but the random effects model model for 3 degreesof freedom is smaller than to 57.04 and the critical value statistic works out

87

this; it is 7.81.

56

The Hausman statistic rejects the null hypothesis for our model (equation

5.3) at the 0.05 level, and hence the conclusion is that the random-effects model is not

be better off using the fixed-effectsmodel in the value-relevance so appropriate we would

tests. Following the value-relevance literature (e.g., Biddle et al. 1995,1997; Chen and Dodd 1997; Worthington and West 2004; Bao, and Bao 1998; Feltham et al. 2004), hypothesis 2 was tested using a two step process. For the first step, the value-relevanceof the three explanatory variables; that is, EVA, TSR, and Tobin's Q was evaluated. each of To accomplish this, each of these three variables was specified as the explanatory variable in separateunivariate regressionswith MVA as the dependentvariable:

In MVA,, = ao + a, In E VAj, + ej, Tobin's Qj, + u,, =A+Aln In MVA,, = yo + y, In TSR,,+ vi, In MVA,

(5.7) (5.8) (5.9)

Value-relevance was, then, assessedby comparing R-squares for the three regressions. A, B, and Q displays the results of the regressionsand shows that EVA Table 5.6 (Panels it possessesthe greatest information power in has the greatest value-relevance as MVA. The main observation is that EVA has a significantly the variation in the explaining Q and TSR. Tobin's Q in return has a larger R-square (39.12%) than that of Tobin's (34.80%) than that of TSR (5.70%). Meanwhile, the higher R-square significantly (EVA, Tobin's Q, TSR) are positive and of the three variables univariate coefficients
from Schmidt (2005), p.405. 56 Chi-squarecritical values are

88

(a=0.3325, t=28.28; 0=0.4931, t=39.63; y=0.1202, t=10.53; respectively), significant indicating that these variables are relatively reliable predictors of shareholdervalue. These further confirmed by the multivariate regression (Equation 5.3) as previously results are

stated.
In the second step, a set of tests was conducted to find out which of the three of shareholder wealth provides value-relevance data beyond that provided by predictors In these tests, each of the three explanatory variables was paired other measures. in a multivariate regression. For example, the incremental alternately with each other for EVA over TSR is obtained from a multivariate regressionwhere both value-relevance EVA and TSR are specified as explanatory variables:

) In MVA,, =, 80 +Aln(EVA,, ) +, 92 ln(TSR,, + e,

(5.10)

Taking the R-square from this pair-wise regression (Equation 5.10), and subtracting the individual R-square for TSR obtained in the earlier univariate regression (Equation 5.9), incremental value-relevance of EVA over TSR. Likewise, the incremental yields the for EVA over Tobin's Q is obtained from a multivariate regressionwhere value-relevance both EVA and Tobin's Q are specified as explanatory variables:

In AfVA,, =A+Aln(EVA,

I)

82 In(Tobin' s Q,,) + e,, +,

(5.11)

89

Taking the R-square from this pair-wise regression (Equation 5.11), and subtracting the

individual R-squarefor Tobin's Q obtained in the earlier univariate regression(Equation 5.8), yields the incremental value-relevanceof EVA over Tobin's Q. Similarly, the incremental value-relevancefor Tobin's Q over TSR is obtained from a multivariate
both Tobin's Q and TSR are specified as explanatory variables: regressionwhere

In MVA,, = 80 + InA(Tobin's

Qj +, 82 ln(TSR,, + e,, )

(5.12)

Taking the R-square from this pair-wise regression (Equation 5.12), and subtracting the individual R-square for TSR obtained in the earlier univariate regression (Equation 5.9), the incremental value-relevance of Tobin's Q over TSR. yields As can be seen in Table 5.6 and 5.7, EVA is the most significant explanatory by itself or when paired with Tobin's Q. The pair-wise combinations that most metric MVA, in order of decreasing power, are EVA/Tobin's Q (52.93%), EVA/TSR explain (42.11%), and Tobin's Q/TSR (35.66%). Once again, as hypothesized, the comparison of evidence of the relative value-relevance of the EVA metric regressions provides strong TSR and Tobin's Q in explaining shareholderwealth. over The results in Table 5.8 Panel C provide incremental value-relevancetests for the EVA, TSR, and Tobin's Q. For example, the incremental valuepair-wise combinations of EVA over TSR (36.41%) can be quantified by comparing the R-squaresof the relevanceof The value-relevance of the pair-wise comparison of EVA and TSR two regressions: from Table 5.8 Panel B minus the value-relevance of TSR (5.70%) from Table (42.11%)

90

5.8 PanelA. Likewise, the incrementalvalue-relevance EVA over Tobin's Q (18.13%) of be calculatedby comparingthe R-squares two regressions: The value-relevance of can of the pair-wisecomparisonof EVA and Tobin's Q (52.93%)from Table 5.8 PanelB minus
the value-relevance of Tobin's Q (34-80%) from Table 5.8 Panel A. The pair-wise

(EVA/ TSR) and (EVA/Tobin's Q) indicate that explanatory power has combinations of increasedby 36.41% and 18.13% respectively over the EVA measurealone. As summarized in Table 5.8, the overall results indicate that EVA exhibits the

largestrelative and incremental information usefulness over TSR and Tobin's Q. These
results convincingly support the claims made by EVA proponents that EVA outperforms in explaining shareholderwealth. other performance measures 5.3.3 Hypothesis Three

"Components unique to EVA help to explain contemporaneous MVA beyond the explanation given by value-based performance measures such as TSR and

Tobin's Q."
To test for the incremental value-relevance of EVA components over the value TSR and Tobin's Q, the following multivariate regressionmodel was used: relevanceof

In MVA,, =A+Aln(profitability +, 82 In(total invested capitalj sPreadj (TSR,,)+ 85 In(Tobin's Qj + e,, 83 In(Lagged EVAJ +, 84 1n +,

(5.13)

91

This model was also estimated using a pooled least square regression. The dependent

MVA for firm (i) in period (t), and the explanatory variable was variables were
Profitability Spread (PS), Total Invested Capital (TIC), Lagged EVA (Lagged EVA), Total

Return (TSR), and Tobin's Q. Profitability Spread,TIC, TSR, and Tobin's Shareholder
defined. Descriptive statistics are displayed in Table 5.1. To account for are as previously in the variables, the values are expressedin natural logarithms. any non-normality As reported in Table 5.9, Panel A, the results of this regression indicate a positive and significant relationship to all of the independentvariables with an overall R-squareof 0.4724. The estimated slope coefficients for profitability spread, TIC, Lagged EVA, TSR, Tobin's Q are 0.1293,0.3064,0.0481,0.0841, and and 0.2506 respectively and all are

5%. Examination of the Variance Inflation Factor (VIF) shows that there is significant at in the regressionmodel. As shown in Table 5.9, Panel no serious multicollinearity problem B, the highest VIF in this model is (3.75). This suggests that, while multicol linearity, it does not create a serious problem. Again as detailed in Table 5.9, the results of exists, this regression not only provided strong evidence of the significance and direction of the hypothesized; but also established a baseline by which to relationships as previously the incremental value-relevance of EVA. analyze Here again, to choose between the fixed or random-effects models, the Hausman (1978) specification test was used. As can be seenfrom Table 5.10, the test statistic works 178.86 and the critical value for 5 degrees of freedom is smaller than this; it is out to

92

Thus, the Hausmanspecificationtest confirms the superiorityof the fixed-effects 11.07.57 the random-effectsmodel. Moreover, as summarizedin Table 5.8, the results model over
indicate a significant relationship to all of the independentvariables and of this regression baseline for analyzing the incremental value-relevanceof EVA components.To provide a this incremental value-relevance question, EVA was broken down into three address that is: Profitability Spread,Total Invested Capital, and Lagged EVA; and the components, MVA was assessed. of each component toward explaining contemporaneous contribution Following the value-relevance methodology, this hypothesiswas tested using a two First, the value-relevance of each of the five explanatory variables; that is, step process. Profitability Spread, Total Invested Capital, Lagged EVA, TSR, and Tobin's Q was To accomplish this, each of these variables was specified as the explanatory evaluated. in separateunivariate regressionswith MVA as the dependentvariable: variable

In MVA,, = oto+ a, In profitability spread,, + e,, InMVA,, =, O+AIn total invested capital,, + e,, In MVA,, = yo + 7, In Lagged EVA,, + e,, In In MVA = 90 + 15, TSR + e In MVA,, = 7ro+ 7r, In Tobin's Q,, + e,,

(5.14)

(5.16) (5.17) (5.18)

from Schmidt (2005), p.405. 57 Chi-squarecritical values are

93

by Value-relevancewas, then, assessed comparing R-squaresof the five regressions.Table 5.11 (Panel A, B, C, D, and E) presents the results and shows that the profitability spread has the greatest value-relevance-that is, it possess the greatest information power in

the variation in the MVA. The main observation was that the profitability explaining has a significantly higher R-square (38.65%) than that of Tobin's Q (34.80%) and spread TSR (5.70%). Tobin's Q in return has a significantly higher R-square (34.80%) than Lagged EVA (30.20%) and TIC (14.09%). In the second step, a set of tests was performed to find out which of the five data beyond that provided by predictors of shareholder wealth provides value-relevance In these tests, each of the five explanatory variables was paired alternately other measures. in a multivariate regression. For example, the incremental value-relevance with each other for the profitability spread over Tobin's Q was obtained from a multivariate regression both the profitability spreadand Tobin's Q were specified as explanatory variables: where

InMVAj, =, go +Aln(profitability

) +, 82 In(Tobin's Qj,) + e,, spread,,

(5.19)

Taking the R-square from this pair-wise regression (Equation 5.19), and subtracting the individual R-square for Tobin's Q obtained in the earlier univariate regression (Equation 5.18), yielded the incremental value-relevance of profitability spread over Tobin's for the profitability spread over TSR was Likewise, the incremental value-relevance both profitability spread and TSR were from a multivariate regression where obtained specified as explanatory variables:

94

In MVA,, = 60 +Aln(profitability

) spreadj +, 82 ln(TSR,, + ei,

(5.20)

Taking the R-square from this pair-wise regression (Equation 5.20), and subtracting the individual R-square for TSR obtained in the earlier univariate regression (Equation 5.17), the incremental value-relevance of profitability spread over TSR. Similarly, the yielded incremental value-relevance for profitability spread (PS) over TIC was obtained from a multivariate regression where both PS and TIC were specified as explanatory variables:

InMVA,,= 80 +A ln(profitability spreadj +,62In(TICj + e,,

(5.21)

Taking the R-square from this pair-wise regression (Equation 5.21), and subtracting the individual R-square for TIC obtained in the earlier univariate regression (Equation 5.15), the incremental value-relevance of profitability spread over TIC. yielded Also, the

incremental value-relevance for profitability spread (PS) over Lagged EVA was obtained from a multivariate regression where both PS and Lagged EVA were specified as explanatory variables:

InMVA,, =, go +Aln(profitability

+, 82 In(Lagged EVAJ + e,, spreadj

(5.22)

95

Taking the R-square from this pair-wise regression (Equation 5.22), and subtracting the

individual R-square for Lagged EVA obtained in the earlier univariate regression
(Equation 5.16), yielded the incremental value-relevance of profitability spread over

Lagged EVA. Similarly, the incremental value-relevance TIC over Tobin's Q was for from a multivariate regressionwhere both TIC and Tobin's Q were specifiedas obtained
explanatoryvariables:

In MVA,, = 80 +AIn(total

invested copitalj +, 82 In(Tobin's Qj + e,,

(5.23)

Taking the R-square from this pair-wise regression (Equation 5.23), and subtracting the individual R-square for Tobin's Q obtained in the earlier univariate regression (Equation 5.18), yielded the incremental value-relevance of TIC over Tobin's Q. Also, the

incrementalvalue-relevance for TIC over TSR was obtained from a multivariate regression both TIC and TSR were specified as explanatory variables: where

In MVA,, = 60 +, 8, In(total invested capitalj +, 82 ln(TSR,, + e,, )

(5.24)

Taking the R-square from this pair-wise regression (Equation 5.24), and subtracting the individual R-square for TSR obtained in the earlier univariate regression (Equation 5.17), incremental value-relevance of TIC over TSR. yielded the Similarly, the incremental

EVA over Tobin's Q was obtained from a multivariate for Lagged value-relevance

96

both Lagged EVA and Tobin's Q regressionwhere were specified as explanatory variables:

In MVAj, =, 80 +Aln(Lagged

EVAJ +, 82 In(Tobin's Qj + e,,

(5.25)

Taking the R-square from this pair-wise regression (Equation 5.25), and subtracting the

individual R-squarefor Tobin's Q obtainedin the earlier univariate (Equation regression

5.18), yielded the incremental value-relevanceof Lagged EVA over Tobin's Q. Similarly, the incremental value-relevance for Lagged EVA over TSR was obtained from a multivariate regression where both Lagged EVA and TSR were specified as explanatory

variables:

In MVA,, = 80 +, 81 1n aggedE VA,, 82 ln(TSR,, + e,, (L )+ )

(5.26)

Taking the R-square from this pair-wise regression (Equation 5.26), and subtracting the individual R-square for TSR obtained in the earlier univariate regression (Equation 5.17), yielded the incremental value-relevance of Lagged EVA over TSR. Finally, the incremental value-relevance for TIC over Lagged EVA was obtained from a multivariate regression where both TIC and Lagged EVA were specified as explanatory variables:

97

In MVA,, =, 80 +Aln(total

invested capiral,, ) +, 82 In(Lagged EVAJ + e,,

(5.27)

Taking the R-square from this pair-wise regression (Equation 5.27), and subtracting the individual R-square for Lagged EVA obtained in the earlier univariate regression (Equation 5.17), yielded the incremental value-relevanceof TIC over Lagged EVA. Also,

for the incrementalvalue-relevance TSR over Tobin's Q was obtainedfrom a multivariate

both TSR and Tobin's Q were specified as explanatory variables: regressionwhere

In MVA,, = flo +Aln(TSRi,

) +, 82 In(Tobin's Qj + e,,

(5.28)

Taking the R-square from this pair-wise regression (Equation 5.28), and subtracting the individual R-square for Tobin's Q obtained in the earlier univariate regression (Equation 5.18), yielded the incremental value-relevanceof TSR over Tobin's Table 5.12 (Panel A, B, C, D, E, F, G, H, 1, and J) displays the results of the pairthe EVA components. As summarized in Table 5.13 Panel A and B, wise regressions of (PS), one of the primary drivers of EVA, is indeed the most the profitability spread by itself (38.65%) or when paired with Tobin's Q (43.11%). significant explanatory metric The pair-wise combinations that most explain MVA, in order of decreasingexplanatory Q (43.11%), Lagged EVA/Tobin's Q (42.77%), PS/Lagged EVA power, are PS/Tobin's PS/TSR (37.94%), TIC/Tobin's Q (37.84%), TSR/Tobin's (40.30%), PS/TIC (40.29%), TIC/Lagged EVA (28.44%),and TIC/TSR (35.66%), Lagged EVA./TSR (31.56%), further evidence of the (15.95%). Here again the comparison of regressions provides

98

EVA over TSR and Tobin's Q in explaining of relativevalue-relevance shareholder wealth

hypothesised. as Furthermore, Table 5.14 presentsthe incremental value-relevance results. In fact, it the hypothesis tests of Equation (5.13) that profitability spread (PS), total provides invested capital (TIC), lagged EVA (Lagged EVA), TSR, and Tobin's Q have equal incremental value-relevance, that is, equal explanatory power. Taking the R-square from in Table 5.13 Panel B, and subtracting the R-squareobtained in eachpair-wise regression the earlier univariate regression (Table 5.11) yielded the tests for incremental valueFor example, in Table 5.14 ProfitabilitySpread / TSR (32.24%) is equal to the relevance. relative value-relevance of the pair-wise comparison of ProfitabilitySpread and TSR (37.94%) from Table 5.13 Panel B minus the individual value-relevanceof TSR (5.70%) from Table 5.13 Panel A. Thus, the explanatory power has increasedby 32.24% due to the ProfitabilitySpread measure. Finally as hypothesised, value-relevance tests once again MVA to be more closely associated to EVA and its components than TSR or reveal

Tobin's

5.4 Concluding Remarks

findings and analysis and presents a detailed Chapter 5 combines the research by comparing them with the main ones discussedin the critical discussion of these results data of 12,000 firm-year observationstaken from literature review. Using a sample of panel files over the period 1991databaseand the clatastream the Stem Stewart 1000 EVA/MVA is a function of 2002, this study has found compelling evidence that shareholder value

99

EVA. It also substantiates other evidence that EVA outperforms other traditional in explaining the cumulative change in shareholderwealth. Valueperformance measures tests reveal EVA to be more highly associatedwith shareholderwealth than TSR relevance Tobin's Q. and Incremental tests have also suggested that EVA possesses the largest

(or information usefulness)over TSR and Tobin's Q. All these results explanatory power the claims made by EVA proponents of EVA's usefulness as an conclusively support internal and external corporate performance measurement.

100

CHAPTER 6 CONCLUSIONS, LIMITATIONS, RESEARCH

This chapter summarizes the ma or findings of this thesis, discusses its j identifies research limitations, as well; it offers some thoughts about contributions, and for future research. opportunities Over the last two decades,a dramatic change has occurred in the way corporations Globalization, technological changes, ownership concentration, are run or operate. the information revolution including the internet, as well as the ever accountability, increasing sophistication of the financial markets have become the primary forces behind the transformation of corporations and the climate in which they operate. Companies the globe are under great pressurenot only to adapt to this new climate, but also to around in all markets in which they compete - namely, the product perform consistently well labour market, and the capital market. If they do not adapt on all levels; they market, the be forced to go bankrupt or will have face the threat of being taken over. will All the above mentioned, but in particular, the increasein ownership concentration hands of institutional investors have set the stage for increased of common shares in the to refocus their efforts toward shareholderconcerns. pressure on corporate management Today, academics, business professionals, and stock market analysts widely agree that financial objective of a business is the most vital maximizing shareholder wealth

AND FUTURE

101

However, widely divergent opinions exist as to how this value enterprise. can be identified, measured,and ultimately, optimized. For awhile now, the traditional accounting measuresof corporate performancesuch as ROA, ROE, and EPS are meeting up with ever increasing criticism and dissatisfaction. Opponentsargue that these measuresprovide a relatively poor guide to shareholder value. Rappopart's pioneering work (1986,1998) that focused on shareholder value took into account the shortcomings of the traditional accounting measures,thus preparing the way for a value-based management (VBM) approach. This new approach has gained

it outlines two important propositions: first, that shareholdervalue widespreadapproval as is the primary corporate objective, and secondly, that economic income of a creation by its EVA, is the primary measure of corporate performance company, as expressed (Davies, 2000, p. 38). Shareholder value has thus long been the theme of the financial economist. Studies show that companies thrive when real economic value is created for their shareholders (e.g., McKinsey et al., 2005; Copeland et al., 1996; Rappopart, 1994,1998; McTaggert et 1994). Yet, this can only happen if corporations are capable of making strategic and al., decisions that will yield a return in excessof their cost of capital. If they are operating there is an added benefit to society. Therefore, capable of carrying this out; of course, be part of any company's culture. These fundamental principles value creation should have been around for a long time, and the events of the recent concerning value creation in them. When managers,boards of directors, have only strengthened our conviction past be detrimental to the overall can investors forget these simple truths, the consequences and

102

operation of the company and society, in general. Let's look back over the last 30 or so Consider the 1970s, with the rise and fall of big business years. conglomerates; the 80s the hostile takeovers in the United States;the 90's with the collapse of Japan'sbubble with

economy and the SoutheastAsian crisis; the Internet bubble breaking; as well as the
corporate governance scandals; and last, but not least, consider the ongoing mortgage

crisis.
Performance measurement, like many other subfields in corporate finance and has had numerous controversies and debates on issues management accounting, its theory, its methodologies, as well as the implementation of its findings. surrounding The issue of whether Economic Value Added (EVA), a variant of residual income, is highly correlated with the market value of the firm as its proponentsclaim has beenmired in ongoing theoretical and methodological controversy. The primary motivation for this from the desire to Provide empirical evidence for the value-relevanceof study stemmed EVA vis-a-vis other value-based performance metrics in predicting shareholderwealth. The secondary motivation was due to the recent interest in the theory and application of by academics, consulting firms, money managers,business value-based metrics shown in the context of the ever growing prominent analysts,and corporate executives evidenced finance and accounting conferences,publications, and other professional meetings. draws upon extensive prior research to develop three interrelated The study hypotheses.First, I hypothesize that EVA is positively and significantly relatedto research

103

58Second, I hypothesize Market Value Added (MVA). that EVA is more highly associated MVA than other commonly-used value-basedperformancemeasures with such as TSR and Tobin's Q. Third, I hypothesize that components unique to EVA help in explaining

MVA beyondthat explainedby TSR andTobin's contemporaneous

Contrary to what other studies (e.g., Biddle et al. 1997; Chen & Dodd 1997) the findings of this study show that EVA dominates value-basedperformance concluded, such as TSR and Tobin's Q in its association with shareholderwealth. In this metrics EVA, TSR, and Tobin's Q were treated as competing value-basedperformance study, measuresand a multivariate regression model was run using a sample of panel data of 12,000firm-year observations covering the period 1991-2002obtained from Stem Stewart DATASTREAM as described in the preceding chapters. The value-relevancetests and revealed that shareholder value (or MVA) is indeed more closely associatedwith EVA than TSR and Tobin's Q. The analysis of these regressionsindicates that the relationship betweenEVA and MVA. is significant and that EVA exhibits the largest explanatorypower the measures. In fact, EVA was significant alone in the univariate regressionsand among incrementally value-relevant in combination with TSR or Tobin's Q in the was These results, therefore, persuasively support the claims made by multivariate regressions. EVA advocatesthat it outperforms other performance measuresin explaining shareholder wealth.

(or MVA) representsthe value created over the lifetime of the Market value Added previously stated, for the past and current value of the strategies of the firm. firm and can be seen as a proxy 5' As

104

Overall, the findings in this study are broadly comparable to prior studies the information usefulness of EVA, including O'Byrne (1996), Bao & Bao supporting (1998), Worthington & West (2004), Zafiris and Bayldon (1999), Lehn and Makhija (1997), Grant (2003), and Feltharn et al. (2004), among others. However, the disparity between the findings of this study and that of others requires explanation. It is thought likely that this can be attributed to two major possibilities: o o differences in researchdesign and methodology; and differences in the accounting principles and variables definition.

This study extends prior studies on the relationship between value-based performance metrics and shareholder value creation. The objective of this study was to empirically examine the hypothesis that EVA is highly associatedwith MVA. The purpose of the study, though, was not to fully explain the determinantsof MVA, but only to show how well EVA acts as a genuine explanatory variable for MVA, in order to justify its for performance measurement, CEO compensation, financial reporting, appropriateness Two more commonly used value-based performance and shareholder value creation. is, Total Shareholder Return (TSR) and Tobin's Q were also consideredto metrics-that highlight the value-relevance of EVA vis-a-vis these measuresin predicting shareholder

wealth.
as This study will contribute to the growing literature on performancemeasurement it made use of pooled time-series cross-sectionaldata, which certainly allows for greater EVA. Moreover, the current study is the first study empirical certainty on the usefulnessof data on four value-basedperformance metrics-namely, EVA, of which I am aware to use

105

MVA, TSR and Tobin's Q and covers a more recentperiod in the contextof corporate
performanceand shareholder value creation. Thus, this study provides evidencethat would

prove useful to policy makers who are interested in EVA as a replacement or a complementto traditional accounting-basedperformancemetrics for their decision-making and compensationpurposes. However, this study is not without some limitations. Initially, it intendedthat it was would consist of all the five prominent value-basedmetrics-that is, CFROI, SVA, TSR, Tobin's Q, and EVA. But due to a lack of data, the first two metrics were droppedfrom the limiting it to just three performance metrics. Also, it was not the study, goal of this study to how firms achieved their levels of EVA or to detail how firms should explore go about increasingtheir EVA level to deliver higher levels of market value added. Such research would be dramatically essential, but is beyond the scopeof the measures publicly available at this time. The results presented in this thesis have, though, provided some necessary insight into the value-relevance of EVA vis-a-vis commonly-used value-based

performancemeasures.However, several other issueswould have to be consideredfurther in order to develop a more complete understandingof what factors drive shareholder value. Rather than replicating current studies with slightly different settings, future research be expanded to include more financial and non-financial metrics such as customer should loyalty, R&D, employee satisfaction, IT, and productivity measures,to mention only a few.

106

REFERENCES
Ahn, H., (2001). Applying the Balanced Scorecard concept: an experience report, Long Range Planning. Vol. 34, No. 4, pp.441-46 1. Alexander, C., (200 1). Market Models: A Guide to Financial Data Analysis. Chichester: Wiley. Anand, M., B. S. Sahay, and S. Saha, (2005). Balanced Scorecard in Indian Companies, VIKALPA. Vol. 30, No. 2, pp. 11-25. Arnold, G. and Matt Davies (ed), (2000). Value-Based Management. Chichester:Wiley. Arnold, G., (2005). Corporate Financial Management. 3rd edition. Harlow, England: Financial Times/Prentice Hall. Ax, C. and T. Bjornenak, (2005). Bundling and Diffusion of Management Accounting Innovations-the case of the balanced scorecard in Sweden, Management Accounting Vol. 16, No. 1, pp. 1-20. Research. Baltagi, B., (2001). Econometric Analysis of Panel Data. Chichester:Wiley. Bao, D., B. Bao, (1998). Usefulness of Value Added and Abnormal Economic Earnings: An Empirical Examination. Journal of Business & Accounting-Vol. 25, No. I and 2, pp. 251-264. Barth, M. E., W. H. Beaver, J.M. Hand, and W.R. Landsman,(1999). Accruals, CashFlows, Equity Values. Review ofAccounting Studies-Vol. 4, No. 3 and 4, pp. 205-220. and Belsley, D.A., (199 1). Conditioning Diagnostics. New York: Wiley. Belsley, D. A., E. Kuh, and R.E. Welsch, (1980). Regression Diagnostics: Identifying Influential Data and Sources of Collinearity, New York: Wiley. F. Siegel, (1995). Relative versus Incremental Information Biddle, G.C., G.S. Seow, and A. Vol. 12, No. 1, pp. 1-23. Research. Content. Contemporary Accounting Wallace, (1997). Does EVA beat Earnings?Evidence Biddle, G.C., R.M. Bowen, and J.S. Firm Values. Journal of Accounting and Associations with stock Returns and on Economics.Vol. 24, pp.301-336.

107

Black, A., P. Wright, and J. Davies, (2001). In Search of Shareholder Value. London: PrenticeHall. Blum, Marshall, and Irwin Friend, (1973). A New Look at the Capital Asset Pricing Model. Journal of Finance, March, pp. 19-27. Bomholt, Graham, (2007). "Extending the Capital asset Pricing Model: the reward beta Accounting and Finance, Vol. 47, pp.69-83. approach", Bromwich, M. and M. Walker, (1998). Residual Income: Past and Future, Management Accounting Research, Vol. 9, pp.391-419. Bruner, R.F., K. M. Eades,R. S. Harris, and R.C. Higgins (1998). "Best Practicesin Estimating the Cost of capital: survey and Synthesis",Financial Nactice and Education, Spring-Summer,pp. 14-28. Chen, S. and J. Dodd, (1997). Economic Value Added (EVA): An Empirical Examination New Corporate Performance Measure. Journal of Managerial Issues, Vol. 9, No.3, of a 318-333. pp. Chung, K. and S.W. Pruitt, (1994). A Simple Approximation of Tobin's Q". Financial Management. Vol. 23, No. 3, pp.70-74. Cooper, S., (2000). Shareholder Wealth or Social Welfare: A StakeholderPerspective,in Value-based Management: Context and Application, Edited by Arnold and Davies, Chichester:Wiley. pp. 81-82. Copeland,T., T. Koller, and J. Murrin, (1996). Valuation: Measuring and Managing the Value of Companies. New York: Wiley. Damodaran,A., (2001). Corporate Finance: Theory and Practice. New York: Wiley. Damodaran on Valuation: security Analysis for Investment Damodaran, A., (2006). Corporate Finance. Hoboken, New Jersey:Wiley. and in Practice: A Critical Review, in ValueDavies, M., (2000). Value-based Management Edited by Arnold and Davies, based Management: Context and Application, Chichester:Wiley. pp. 37-58. Dielman, T., (2005). Applied Brooks/Cole Thomson. Analysis. 4'h Edition. Belmont, CA: Regression

108

Dougherty, C., (2007). Introduction University Press.

to Econometrics, Oxford / New York: Oxford

Drucker, P., (1995), The Information Executives Truly Need. Harvard Business Review. January/February,pp.54-62. Ehrbar, A., (1998). EVA: The Real Key to Creating Wealth. New York: Wiley. Erasmus, P.D. and I.J. Larnbrechts, (2006). EVA and CFROL A Comparative Analysis. Management Dynamics. Vol. 15, No. 1, pp. 15-26. Fabozzi, F.J. and J.L. Grant, (2000). Value-Based Metrics: Foundations and Practice. New Hope, Pennsylvania: Frank J. Fabozzi Associates. Fama, E.F., (1965). The Behavior of Stock Market Prices. Journal of Business. January, pp.34-106. Fama, E.F. and K. R. French, (1992). The Cross-Section of Expected Stock Returns. Journal of Finance. Vol. 47, No.2, pp.427-465. Fama, E.F. and K. R. French, (1993). Common Risk Factors in the Returns on Stocks and Bonds. Journal of Financial Economics. Vol. 33, No. 1, pp.3-56. Fama, E.F. and K. R. French, (1995). Size and Book-to-Market Factors in earnings and 5. Returns.Journal of Finance. Vol. 50, No. 1, pp. 131-15 Fama, E.F. and K. R. French, (1996). Multifactor Explanationsof Asset Pricing Anomalies. Journal of Finance. Vol. 5 1, No. 1, pp.55-84. Farna, E.F. and K. R. French, (2004). The Capital Asset Pricing Model: Theory and Evidence. Journal of Economic Perspectives.Vol. 18,No. 3, pp.25-46. Feltham, G., G, Isaac, C. Mbagwn, and G. Vaidyanathan,(2004). PerhapsEVA Does Beat Earnings-Revisiting Previous Evidence. Journal ofApplied Corporate Finance. Vol. 16, No. 1, pp. 83-88. S. Yu, (2005). Does economic value added (EVA) improve Ferguson, R., J. Rentzler, and Journal ofApplied Finance. Vol. 15,No.2, pp. 101-113. stock performance profitability? Femandez, P., (2002). Valuation Amsterdam: Academic Press. Methods and Shareholder value Creation.

109

Finegan, P.T., (1989). "Financial Incentives Resolve the shareholder-Value Puzzle", Corporate Cashflow. October, pp. 27-32. Friedman, M., (1970). The social responsibility is to increaseits profits. New York Times Magazine. September,pp.32-33. Graham,John R. and Campbell R. Harvey (2001). "The Theory and Practice of Corporate Finance:Evidence from the Field", Journal of Financial Economics, Vol. 60, No.2 and 3, 187-243. pp. Grant U., (1996). Foundations of EVA for investment managers. The Journal of Portfolio Management, Vol. 23, Fall, pp. 41-48. Grant, J., (2003). Foundations of Economic Value Added. Hoboken, New Jersey:Wiley. Greene,H. William, (1993). Econometric Analysis. New York: Macmillan. (2008). Econometric Analysis. 6th edition. Upper Saddle River, NJ: Greene,H. William, PrenticeHall. Econometrics. 4thedition. Boston: McGraw Hill. Gujarati, N. Damodar, (2003). Basic J. (1978). Specification Tests in EconometricsEconometrica, Vol. 46, pp.1251Hausman, 71. W. James,(2000). Linking Balance ScorecardMeasuresto Size and Market Hoque, Z. and Journal of Management Accounting Factors: impact on organisational performance. Vol. 12, pp. 1-17. Research. Panel Data. Cambridge: Cambridge University Press. Hsiao, C., (2003). Analysis of did=46695540I &sid= I &Fmt--4&clientld= I 0843&RQT= http://proquest-umi.com/pqdweb? 309&VName=PQD http://www. valuationissues.com/valiss/servlet/viewarticle?articlelD=95 F. Larcker, (1998). Innovations in Performance Ittner, Christopher D., and David Journal of Accounting Research-Vol. Trends and Research implications. Measurement: 10,pp.205-238. The Balance Scorecard Measures That Drive Kaplan, R. S. and D. P. Norton, (1992). Review. January-February,pp. 71-79. Performance.Harvard Business

110

Kaplan, R. S. and D. P. Norton, (1993). Putting the balanced scorecardto work, Harvard Business Review. September-October,pp. 144-147. Kaplan, R. S. and D. P. Norton, (1996a). Using the Balanced Scorecard as a Strategic ManagementSystem. Harvard Business Review. January-February,pp.75-85. Kaplan, R. S. and D. P. Norton, (1996b). The Balanced Scorecard: Translating Strategy into Action. Harvard Business School Press,Boston, MA. Kaplan, R. S. and D. P. Norton, (2001). The Strategy-focused Organization: How Balanced Scorecard Companies Thrive in the New Business Environment, Harvard BusinessSchool Press,Boston, MA. Kramer, J. and G. Pushner, (1997). An Empirical Analysis of Economic Value Added as a Proxy for Market Value Added. Financial Practice and Education. Spring/Summer, 41-49. pp. Lang, L., R. Stulz, and R. Walking, (1989). Managerial performance, Tobin's q, and the from successful tender offers. Journal of Financial Economics. Vol. 24, pp. 137-154. gains Lehn, K. and A. Makhija, (1996). EVA and MVA as Performance Measuresand Signals for Strategic Change. Strategy and Leadership. May/June, pp.34-38. Lehn, K. and A. Makhijia, (1997). EVA, Accounting Profits, and CEO Turnover: An Empirical Examination, 1985-1994.Journal ofApplied Corporate Finance. Vol 10,No.2, 90-97. pp. Lindenberg, E. and S. Ross, (1981). Tobin's Q Ratio and Industrial Organization. Journal Business. Vol. 54, pp. 1-32. of Valuation of Risk Assets and the Selection of Risky Investmentsin Lintner, J., (1965). The Stock Portfolios and Capital Budgets. Review of Economics and Statistics. pp. 13-37. Livant, J. and P. Zarowin, (1990). The Incremental Information Content of Cash-Flow Components.Journal ofAccounting and Economics. Vol. 13,No. 1, pp. 25-46. Valuation: A Total Systems Approach to Valuing the Madden, B. J., (1999). CFROI Firm. Woburn, N4A: Butterworth-Heinemann. R. Reilly, (2003). Beyond the Balance Scorecard: Maltz, A. C., A. J. Shenhar, and R. Measures,Long Range Planning, Vol. 36, Refining the Search for Organizational Success No.2, pp. 187-204.

Management Today, (March 1997), p.48. Markovitz, H., (1952). Portfolio Selection. Journal of Finance. pp.77-91. Marshall, A., (1890). The Principle of Economics. London. New York: Macmillan Press. McKinsey & Company, T. Koller, M. Goedhart, D. Wessels, (2005). Valuation: Measuring and Managing the Value of Companies. Hoboken, New Jersey:Wiley. McTaggart, J.M., P.W. Kontes, and M. C. Mankins, (1994). The Value Imperative. New York: Free Press. Milunovich, S. and A. Tsuei, (1996). EVA in the Computer Industry. Journal of Applied Corporate Finance. Vol. 9, No. 1, p. 104-115. Modigliani, F. and Miller, M., (1961). Dividend Policy, Growth and the Valuation of Journal of Business. October pp.411-433. Shares. Morin, R. and S. Jarrell, (2001). Driving Shareholder Value. New York: McGraw-Hill. Myers, R., (1996). Keeping Score: Metric Wars. CFO. October pp.41-50. Neely, A., M. Kennerley, and V. Martinez, (2004). Does the balancedscorecardwork: an investigation, In (Eds, Neely, A., Kennerley, M., and Walters, A. ), Performance empirical Measurement and Management: Public and Private, Cranfield School of Management, Edinburgh, UK, pp.763-777. Norreklit, H., (2000). The Balance on the Balanced Scorecard-a critical analysis of some its assumption,Management Accounting Research, Vol. 5, pp. 289-299. of O'Byrne, S.F. (1996). EVA and Market Value. Journal of Applied Corporate Finance. Vol. 9, No. 1, p. 116-125. O'Hanlon, J. and K. Peasnell, (1998). Wall Street's Contribution to Management Accounting: The Stem Stewart EVA Financial Management System, Management Accounting Research, Vol. 9, pp.421-444. OhIson, J., (1995). Eamings, Book Values and Dividends in Security Valuation. Contemporary Accounting Research. Vol. 11, pp.661-687. Paulo, S., (2003). Epistemology, ResearchMethodology and Rule 702 of the FederalRules EVA. Journal of Business Ethics. Vol. 44, No. 4, pp.327-341. of Evidence Versus

112

Palliam, Ralph, (2006). Further Evidence on the Information Content of Economic Value Added. Review ofAccounting and Finance. Vol. 5, No. 3, pp.204-215. Perfect, S. and K. Wiles, (1994). Alternative constructions of Tobin's q: An empirical Journal of Empirical Finance. Vol. 1, No. 3, pp.313-341. comparison. Peterson,P. and D. Peterson, (1996). Company Performance and Measures of Value Added. Charlottesville, Virginia: The Research Foundation of the Institute of Chartered Financial Analysts. Pindyck, R. S., and D. L. Rubinfeld, (1998). Econometric Models and Economic Forecasts. Boston: Irwin McGraw-Hill. Rappaport, A., (1998). Creating Shareholder Value: A Guide for Managers and Investors. New York: The Free Press. Reilly, F.K. and K. C. Brown, (1997). Investment Analysis and Portfolio Management. Forth Worth: The Dryden Press. J., (2000). The Cost of Capital and Shareholder Value, in Value-based Rutterford, Management: Context and Application, Edited by Arnold and Davies, Chichester: Wiley. pp. 135-150. Samuel, B., (1996). Capitalism with a Human Face, Fontana Press.

Schmidt, StephenJ., (2005). Econometrics. Boston: McGraw-Hill. date). Creating Value through EN. A.: Myth or Shaked, I., A. Michel, and P. Leroy, (no Reality? http: //www. strategy-business. com/strategy/97404/p A Theory of Market Equilibrium Under Sharpe, W., (1964). Capital Asset Prices: Finance. September,pp.425-442. Conditions of Risk. Journal of New Jersey:Princeton University Press. Shiller, R., (2000). Irrational Exuberance. Capitalism. London: Little, Brown G., (2000). Open Society: Reforming Global Soros, Company. and A Descriptive Analysis on the Implementation Speckbacher,G.J., and T. Pfeiffer, (2003). Countries. Management Accounting Balanced Scorecards in German-speaking of Research. Vol. 14, No. 4, pp. 361-387.

113

Stark, A. W. and H.M. Thomas, (1998). On the Empirical RelationshipbetweenMarket Value and Residual Income in the U.K. Management Accounting Research.Vol.9, 445-460. pp.
Stem, J., (1996). EVA and Strategic Performance Measurement. Global Finance2000: A Handbook of Strategy and Organization, The ConferenceBoard. Stewart,B., (1991). The Quest for Value. New York: Harper Business. Stewart, B., (1994). EVA: Fact and Fantasy.Journal ofApplied Corporate Finance. Vol. 7, Summer,pp. 71-84. Stewart, B., (1995). EVA Works - But Not if You Make These Common Mistakes, Fortune, May 1, pp. 117-118. Studenmund,A. H., (2006). Using Econometrics: A Practical Guide. 5th edition, Boston: Pearson. The Economist, (1997, August 2). Valuing Companies:A Star to SaiI By? pp. 61-63. Titman, S. and J. D. Martin, (2008). Valuation: The Art & Science of Corporate Investment Decisions. Boston: Pearson/AddisonWesley. Tobin, J., (1969). A general Equilibrium Approach to Monetary Theory. Journal of Money, Credit, and Banking. Vol. 1, No. 1, pp. 15-29. Tobin, J. and W. Brainard, (1968). Pitfalls in Financial Model Building. American Economic Review. Vol. 58, No. 2, pp. 99-122. Topkis, M., (1996). A New Way to Find Bargains. Fortune. December9, pp-265-266. Tully, S., (1993). The Real Key to Creating Wealth. Fortune. September20, pp.38-50. Tully, S., (1998). America's GreatestWealth Creators.Fortune. November 9, pp.93-204. Tully, S., (1999). The EVA Advantage, Fortune, March 29, p-210. J. Pettit, (1996). EVA for Banks: Value Creation, Risk Uyemura, D., C. Kantor, and Journal of Applied Corporate Finance. Management, and Profitability Measurement. Vol. 9, No. 2, pp. 94-113. Perfonnance 1000: Using EVA to Build Market Walbert, L., (1995). The Stem Stewart Corporate Finance. Winter, pp. 109-120. Value. Journal ofApplied

114

Wilbur, L. and S.G. Badrinath, (1997). On the Measurement Tobin's q. Journal of of Financial Economics. Vol. 44, pp.77-122;
Wooldridge, J., (2002). Econometrics Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press. Worthington, A. C. and T. West, (2004). Australian Evidence Concerning the Information Content of Economic Value Added. Australian Journal of Management. Vol. 29, No.2, 201-223. pp. Yook, K. C., (1999). Estimating EVA Using Compustat PC Plus. Financial Practice and Education. Vol. 9, No. 2, pp.33-37. Young, D. and S. O'Byrne, (2001). EVA and Value-Based Management. New York: McGraw-Hill. Zafiris, N. and R. Bayldon, (1999). Economic Value Added and Market Value Added: a Version and Application. Journal ofApplied Accounting Research. December,pp. simple 84-107. Zarowin, S., (1995). The Future of Finance. Journal ofAccountancy. Vol. 180,pp. 47-49.

115

APPENDICES

Appendix A: Figures Appendix B: Tables Appendix C: The 2002 Stern Stewart Performance 1000

116

Appendix

FIGURE I THE SHAREHOLDER VALUE ANALYSIS NETWORK

CORPORATE OBJECTIVE

Shareholder Value Added (SVA)

Shareholder Return Dividends Capital Gains

VALUATION COMPONENTS

Cash Flow from Operations

Discount Rate

Debt

VALUE DRIVERS

Value Growth Duration

Sales Growth " Operating Profit " Margin Income Tax Rate "

Working Capital Investment Fixed Capital Investment

Cost of Capital

MANAGEMENT DECISIONS

Investment

Financing

Source: Rappaport (1998), p. 56.

117

FIGURE 2 A TYPICAL FINANCIAL MANAGEMENT FINANCE) SYSTEM (FUZZY

Source: G. Bennett Stewart, 111,Focused Finance. http: //www. valuationissues. com/valiss/serviet/Viewarticle? articlelD=95

118

FIGURE 3
EVA: A Simplified and Focused Financial Management System

Source: Adopted from Morin and Jarrell, P-46

119

FIGURE 4 Balanced Scorecard: Outcome Measures & Performance Drivers

Causal Chain
inand Lagong

Customers

Internal Business Process

Leading Learning and Growth

(2004), &Smith Fletcher Source: P-5

120

Atmendix B:

TABLE 2.1
HOW PERFORMANCE MEASURES "EXPLAIN" MVA
10% 15%-20% 35% about 50%

CHANGES IN
R2

Variable/Model Growth in Sales EPS and NI ROA, ROE, RONA EVA

Source: Stewart (1994), p.75.

121

TABLE 4.1
CALCULATION OF NOPAT FROM FINANCIAL DATA

A. Bottom-up approach Begin: Operating profit after depreciation and amortization Add: Implied interest expenseon operating leases Increasein LIFO reserve Increasein accumulated goodwill amortization Increasein bad-debt reserve Increasein capitalized researchand development Increasein cumulative write-offs of special items Equals: Adjusted operating profit before taxes Subtract: Cash operating taxes Equals: NOPAT B. Top-down roach Begin: Sales Subtract: Cost of goods sold Selling, general, and administrative expenses De reciation Add: Implied interest expenseon operating leases Increasein LIFO reserve Other operating income Equals: Adjusted operating profit before taxes Subtract: Cash operating taxes _Equals: NOPAT III, %qu%', I Note: Table based on information in (j. tienneu 3LCWdFL I 11c; )LJL%JJL York: Harper Collins, 1991).

122

TABLE 4.2
CALCULATION OF CAPITAL USING ACCOUNTING FINANCIAL

STATEMENTS
A. Asset approach Begin Net (short term) operating assets Add: LIFO reserve Net plant and equipment Other assets Goodwill (Intangibles) Accumulated goodwill amortization Presentvalue of operating leases Bad-debt reserve C italized researchand development Cumulative write-offs of special items Equals: Capital A Sourceoffinancing approach Begin: Book value of common equity Add equity equivalents: Preferred stock Minority interest Deferred income tax reserve LIFO reserve Accumulated goodwill amortization Add debt and debt equivalents: Interest-bearing short-term debt Long-term debt Capitalized lease obligations Presentvalue of noncapitalized leases Equals. Capital _ Note: Table based on information in G. Bennett Stewart III, ine yuest ior value klNcw York: Harper Collins, 1991).

123

TABLE 5.1 DESCRIPTIVE STATISTICS OF VARIABLES EMPLOYED IN THE REGRESSION MODELS

Variable MVA overall between within EVA overall between within NOPAT overall between within TSR overall between within TobinQ overall between within SPREAD overall between within TIC overall between within ROIC overall between within WACC overall between within lagEVA overall between within -81.13532 . 0921327 . 0933339 5796.468 . 0015204 8.405437 . 2588763 401.9223 -81.15118 Mean 5266.249 Std. Dev. Min -72667 -9778.25 -167248.8 -29732 -7602 -24906.6 -24849 -3121.9 -21325.18 -. 938144 -. 606695 -4.37565 -6108.15 -500.13 -8664.77 -7.86567 -. 63312 -10.05144 -1395 -534.9167 -51475.03 -7.78067 -. 49786 -9.956298 042 . 05125 . 0383593 . -29732 -7758.75 -25692.14 Max 505162 188466.5 423189.8 8358 4312 9583.849 18723 13073.25 9938.506 44.46667 3.850578 40.87497 49898.73 8626.205 41280.93 21.57341 4.149515 17.42542 233490 116989 172904.1 21.67341 4.227182 17.53956 . . 188 173 T-bar N= n= T-bar = T-bar N= n= = T-bar N= n= = T-bar N= n= = T-bar T-bar N= n= T-bar N= n= = T-bar N= n= = T-bar N= n= = Observations N= n= = 9532 1000 9.532 9578 1000 9.578 9889 1000 9.889 8954 914 9.7965 9497 936 20431.39 14906.85 13329.41 952.4533 665.9038 687.4061 1116.824 1014.2 589.4312 . . . 8658269 3386405 8177131 525.884 284.1693 477.7714 . 3194024 . 2690804 166963

= 10.1464 9578 N= n= = 1000 9.578 10016 1000 10.016 9578 1000 9.578 9845 1000 9.845 9577 1000 9.577

. 13470.59 11586.65 6526.184 . . . . .

3195998 1663764 2695485 0241721 0216499 0092914

. 952.5018 588.7881 741.8563

. 129466

8358 3805 9199.948

Variable Definitions: Value Added; TSR=Total Shareholder MVA=Market Value Added; EVA=Economic EVA Spread;TIC=Total Invested Return; TobinQ=Tobin's Q; SPREAD=Profitability or WACC=Weighted Average Cost of Capital; Capital; ROIC=Retum on Invested Capital; lagEVA=Lagged Economic Value Added

124

TABLE 5.2 CORRELATION BETWEEN VARIABLES This table reports Pearson correlation coefficient between all variables used in the analysis

A mv EVA TSR TobinQ SPREAD TIC lagEVA 0,6102 0.0346 0.1210 0.1683 0.5541 0.0345

EVA

TSR

TobinQ

SPREAD

TIC

lagEVA

-0.0028 0.0488 0.1351 0.8840 0.0148

0.01977 0.0326 5 0.0572 0.0090 0. 0.0622 0.0002 0.0731 0.1550 04

Thesecorrelations are estimated using 9532 observations. SeeTable 5.1 for all variable definitions.

125

TABLE 5.3 HYPOTHESIS (1) UNIVARIATE REGRESSION RESULTS Panel A IsMVA, go +AsE VA, Model: =, + e,, l Statistic

Numberof Observations Variable Dependent Independent Variable R-Squared Coefficients: Constant Std. error 95% C.I. Independent Variable (sEVA) Std. error 95% C.I. t-statistics: Constant Variable (sEVA) Independent Variable Independent p-value: Constant Variable (sEVA) Independent L-

9449 sMVA sEVA 0.3724

3.364659 0.4932272 2.397813/4.331505 10.32568 0.1370467 10.05704/10.59433

6.82 75.34

0.000 0.000

126

TABLE 5.3 (continued) HYPOTHESIS (1) UNIVARIATE REGRESSION RESULTS Panel B [sMVA,, 80 Model: =, +AsNOPAT,, Statistic Number of Observations Dependent Variable Independent Variable R-Squared Coefficients: Constant Std. error 95% C.I. Independent Variable (sNOPAT) Std. error 95% C.I. t-statistics: Constant Independent Variable (sNOPAT) p-value: Constant Independent Variable (sNOPAT) 0.000 0.000 4.67 75.65 2.297306 0.4917629 1.33333/3.261281 9.986957 0.1320177 9.72817/10.24574 9473 sMVA sNOPAT 0.3748

127

TABLE 5.4 HYPOTHESIS (2) MULTIVARIATE REGRESSION RESULTS Panel A

Model In MVAj, =, 80 +Aln(EVA,, ) +,82 ln(TSR,, +, 83In(Tobin's Q,,) + ) e,,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error
95% C. I. Independent Variable (InEVA) Std. error 95% C. I. Independent Variable (InTSR) Std. error 95% C. I. Independent Variable (InTobin'sQ) Std. error 95% C. I. t-statistics: Constant Independent Variable (InEVA) Independent Variable (InTSR) Independent Variable (InTobin'sQ) p-value: Constant Independent Variable (InEVA) Independent Variable (InTSR) independent Variable (InTobin'sQ) I 0.000 0.000 0.000 29.41 16.80 7.20 16.34

2606 1nMVA InEVA, InTSR,InTobin'sQ 0.3028

2.945609 0.1001646
2.749169/3.142049 0.241497 0.143445 0.2129177/0.2691817 0.0762908 0.0105902 0.0555216/0.09706 0.2897801 0.0177318 0.255005/0.3245551

128

TABLE 5.4 (continued) HYPOTHESIS (2) MULTIVARIATE REGRESSION RESULTS Panel B COLLINEARITY DIAGNOSTICS

Variable InEVA InTSR InTobin'sQ

VIF 1.24 1.06 1.28

Tolerance 0.8034 0.9392 0.7841

R-Squared 0.1966 0.0608 0.2159

Mean VIF: Condition Number:

1.19 9.1878

129

TABLE 5.5 FIXED vs. RANDOM EFFECTS (HAUSMAN TEST)

Model:
Fixed Effects:

MVAj, =, 80 +A(In

(Equafinn 9 EVA,, ) +, 02 (In TSR,,) +, 83 (In Tobin's

Al

+ e,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 0.000 29.41 16.80 7.20 16.34 2.945609 0.1001646 2.749169/3.142049 0.241497 0.143445 0.2129177/0.2691817 0.0762908 0.0105902 0.0555216/0.09706 0.2897801 0.0177318 0.255005/0.3245551 2606 1nMVA InEVA, InTSR, InTobin'sQ 0.3028

130

TABLE 5.5 (continued)

Random Effects:

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. z-statistics: Constant IndependentVariable (InEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 0.000 37.63 21.07 7.43 23.61 3.165228 0.084107 3.000381/3.330075 0.2760608 0.0131028 0.2503798/0.3017417 0.0767928 0.0103314 0.0565435/0.097042 0.3218157 0.0136327 0.2950961/0.3485352 2606 InMVA InEVA, InTSR,InTobin'sQ 0.5401

131

TABLE 5.5 (continued)

Coefficients:
(b) eqFIX 2410497 0762908 2897801 (B) 2760608 0767928 3218157 (b-B) Difference . . . 0350111 -. 000502 -. -. 0320356 sqrt(diag(V_b-V_B)) S. E. . 0023268 0113387 003838

insEVA lnTSR lnsTobin'sQ

. . .

. .

b consistentunder Ho and Ha B inconsistentunder Ha, efficient under Ho Test: Ho: difference in coefficients not systematic (3) chi2 Prob>chi2 (b-B)'[(V b-V B)^(-I)](b-B) 57.04 0.0000

132

STATISTICAL

TABLE 5.6 RESULTS OF STEP ONE OF HYPOTHESIS (2) TESTING Panel A I Model: In MVA,, = ao + a, In EVA,

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InEVA) Std. error 95% C.I. t-statistics: Constant Independent Variable (InEVA) p-value: Constant Independent Variable (InEVA) 0.000 0.000 34.10 28.28 1.410927 0.0413741 1.329807/1.492046 0.332561 0.0117604 0.3095033/0.3556188 4383 lnMVA InEVA 0.3912

133

STATISTICAL

TABLE 5.6 (continued) RESULTS OF STEP ONE OF HYPOTHESIS (2) TESTING Panel B

Modell
Statistic

In MVA,, ---,: +Aln Tobin's Q,, flo

Number of Observations Dependent Variable Independent Variable R-Squared Coefficients: Constant Std. error 95% C.I. Independent Variable (InTobin'Q) Std. error 95% C.I. t-statistics: Constant Independent Variable (InTobin'sQ) p-value: Constant Independent Variable (InTobin'sQ)

6715 lnMVA InTobin's Q 0.3480

2.835 0.0753512 2.687584/2.983016 0.4931501 0.0124448 0.4687538/0.5175465

37.63 39.63

0.000 0.000

134

STATISTICAL

TABLE 5.6 (continued) RESULTS OF STEP ONE OF HYPOTHESIS (2) TESTING Panel C In MVA,, yo + y, In TSR,,+ VII Model:

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant Independent Variable (InTSR) p-value: Constant Independent Variable (InTSR) 0.000 0.000 6.47 10.53 0.1249479 0.0193188 0.0870726/0.1628232 0.1202252 0.0114126 0.0978502 4966 lnMVA InTSR 0.0570

135

TABLE 5.7 STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (2) TESTING Panel A I In MVA,, =, go +Aln(EVA,, ) + #2 ln(TSR,,) Model

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant Independent Variable (InEVA) IndependentVariable (InTSR) p-value: Constant Independent Variable (InEVA) Independent Variable (InTSR) 0.000 0.000 0.000 30.34 22.43 8.45 1.528633 0.0503877 1.42982/1.627445 0.3165868 0.0141168 0.288903/0.3442705 0.0899577 0.0106514 0.0690698/0.1108457 2877 1nMVA InEVA, InTSR 0.4211

136

STATISTICAL

TABLE 5.7 (continued) RESULTS OF STEP TWO OF HYPOTHESIS (2) TESTING Panel B

Model In MVA,, =, 80 +Aln(EVA,, ) +, 82 In(Tobin's Q,,) + e,, Statistic

Number of Observations Dependent Variable Independent Variable R-Squared Coefficients: Constant Std. error 95% C.I. Independent Variable (InEVA) Std. error 95% C.I. Independent Variable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant Independent Variable (InEVA) Independent Variable (InTobin'sQ) p-value: Constant Independent Variable (InEVA) Independent Variable (InTobin'sQ) 0.000 0.000 0.000 34.78 19.38 20.85 2.909641 0.0836618 2.7456/3.073683 0.2405591 0.0124102 0.2162255/0.2648927 0.3223923 0.0124102 0.2162255/0.2648927 3648 lnMVA InEVA, InTobin'sQ 0.5293

137

STATISTICAL

TABLE 5.7 (continued) RESULTS OF STEP TWO OF HYPOTHESIS (2) TESTING Panel C

Model: In MVA,, =, 80 + InA(Tobin's Q,,) +, 82 ln(TSRt) + e,

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTobn'sQ) Std. error 95% C.I. Independent Variable (InTSR) Std. error 95% C.I. t-statistics: Constant Independent Variable (InTobin'sQ) Independent Variable (InTSR) p-value: Constant IndependentVariable (InTobin'sQ) Independent Variable (InTSR) 0.000 0.000 0.000 29.29 28.16 8.55 2.835656 0.0968174 2.645834/3.025478 0.4528348 0.0160809 0.4213063/0.4843633 0.0940835 0.0110063 0.0725042/0.1156627 4410 1nMVA InTobin'sQ, InTSR 0.3566

138

TABLE 5.8 HYPOTHESIS (2) RELATIVE AND INCREMENTAL VALUE-RELEVANCE TESTS

Panel A Relative Value-Relevance Test (individual) EVA Tobin's Q TSR 0 -- - - 39.12% > 34.80% > 5.70%

Panel B Relative Value-Relevance Test (pair-wise combinations) EVA/Tobin's Q R2 52.93% > EVA/TSR 42.11% Tobin's Q /TSR 35.66% >

Panel C Incremental Value-Relevance Test EVA/TSR - 36.41% FF EVA/Tobin's 18.13% >

139

TABLE 5.9 HYPOTHESIS (3) MULTIVARIATE REGRESSION RESULTS Panel A In MVA,, =& +Aln(profitability spreadj +A In(total invested copitalj Model +, 83 In(Lagged EVAJ )+ 85 +, 841n(TSR . In(Tobin's Qj + e,, it Statistic Number of Observations DependentVariable Variable Independent 1979 1nMVA InSPREAD,LnTIC, InLagEVA, InTSR, InTobinsQ R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. 3.491256 0.1616931 3.174074/3.808437 0.129359 0.01586 0.0982846/0.1605073 0.3064579 0.0482726 0.2117649/0.4011508 0.04811 0.0178984 0.0130001/0.08322 0.0841718 0.0121273 0.0603826/0.107961 0.4724

140

IndependentVariable (Tobin'sQ) Std. error 95% C.I. t-statistic Constant Variable (InSPREAD) Independent Variable (InTIC) Independent Variable (InLagEVA) Independent Variable (InTSR) Independent IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InSPREAD) IndependentVariable (InTIC) IndependentVariable (InLagEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ)

0.2506594 0.0251512 0.2013222/0.2999966

21.59 8.16 6.35 2.69 6.94 9.97

0.000 0.000 0.000 0.007 0.000 0.000

141

TABLE 5.9 (continued)

HYPOTHESIS (3) MULTIVARIATE REGRESSION RESULTS Panel B Collinearity Diagnostics

Variable (InSPREAD) (InTIC) (InLagEVA) (InTSR) (InTobin'sQ) VIF 3.75 1.13 1.63 1.07 2.94 Tolerance 0.2666 0.8867 0.61340.9380 0.3406 R-Squared 0.7334 0.1133 0.3866 0.0620 0.6594

MeanVIF: ConditionNumber:

2.10 25.3668

142

TABLE 5.10 FIXED vs. RANDOM EFFECTS (Equation 5.13): Hausman Test
In MVAj, =, 80 +Aln(profitability spread,, ) +A In(total invested capital,, ) Model: +A In(Lagged EVA,, ) +, 84 ln(TSR,, ) + 8, In(Tobin's Qj + e,,

Fixed-Effects: Statistic Number of Observations DependentVariable IndependentVariable 1979 lnMVA InSPREAD,LnTIC, InLagEVA, InTSR, InTobinsQ R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. 3.491256 0.1616931 3.174074/3.808437 0.129359 0.01586 0.0982846/0.1605073 0.3064579 0.0482726 0.2117649/0.4011508 0.04811 0.0178984 0.0130001/0.08322 0.0841718 0.0121273 0.0603826/0.107961 0.4724

143

IndependentVariable (Tobin'sQ) Std. error 95% C.I. t-statistic Constant IndependentVariable (InSPREAD) IndependentVariable (InTIC) IndependentVariable (InLagEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InSPREAD) IndependentVariable (InTIC) IndependentVariable (InLagEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ)

0.2506594 0.0251512 0.2013222/0.2999966

21.59 8.16 6.35 2.69 6.94 9.97

0.000 0.000 0.000 0.007 0.000 0.000

TABLE 5.10 (continued)

144

Rai

irl

PA.

loom

]MIUCts'.

Statistic Number of Observations DependentVariable IndependentVariable 1979 lnMVA InSPREAD,LnTIC, InLagEVA, InTSR, InTobinsQ R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. IndependentVariable (Tobin'sQ) Std. error 95% C.I. z-statistic Constant 27.69 3.639706 0.1314526 3.382063/3.897348 0.1187431 0.0147294 0.089874/0.1476122 0.307166 0.0472233 0.2146101/0.3997219 0.1137458 0.0164276 0.0815484/0.1459432 0.0839836 0.0118733 0.0607124/0.1072548 0.2802379 0.0201392 0.2407659/3.897348 0.4951

145

IndependentVariable (InSPREAD) IndependentVariable (InTIC) IndependentVariable (InLagEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InSPREAD) IndependentVariable (InTIC) IndependentVariable (InLagEVA) IndependentVariable (InTSR) IndependentVariable (InTobin'sQ)

8.06 6.50 6.92 7.07 13.92

0.000 0.000 0.000 0.000 0.000 0.000

146

TABLE 5.10 (continued)

Coefficlen s (b) lnSPREAD lnlagEVA lnTSR eqFIX 1293959 (B) 1187431 (b-B) Difference 0106529 . 0656358 -. 0001882 . b-V B)) sqrt(diag(V S. E. 0058808 . 0107 0071054 . 002469 . 50663

04811 . 0841718 .

1137458 . 0839836 .

b consistentunder Ho and Ha inconsistentunder Ha, efficient under Ho B Test: Ho: difference in coefficients not systematic (b-B)'[(V b-V B)/'(-I)I(b-B) 178.86 0.0000 Prob>chi2 = (5) chi2

147

TABLE5.11 STATISTICAL RESULTS OF STEP ONE OF HYPOTHESIS (3) TESTING Panel A IIn MVA,, Model ao + a, In profitability spread,,+ e,, =

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InSPREAD) p-value: Constant IndependentVariable (InSPREAD) 0.000 0.000 31.68 28.88 2.867663 0.0905222 2.690183/3.045142 0.239306 0.0082856 0.2230611/0.2555509 4431 lnMVA InSPREAD 0.3103

148

TABLE 5.11 (continued) STATISTICAL RESULTS OF STEP ONE OF HYPOTHESIS (3) TESTING Panel B
IIn MVA,, 8,, Model =, +Aln total invested capital,, + e,,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTIC) p-value: Constant IndependentVariable (InTIC) 0.000 0.000 -28.97 31.45 -0.3355091 0.011583 -0.3582152/-0.3128031 1.042992 0.0331628 0.9779834/1.108 8432 1nMVA InTIC 0.1409

149

TABLE 5.11 (continued) STATISTICAL RESULTS OF STEP ONE OF HYPOTHESIS (3) TESTING Panel C lIn MVA,, Model: Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InLagEVA) p-value: Constant IndependentVariable (InLagEVA) 0.000 0.000 14.43 10.23 0.7036367 0.0487725 0.6080079/0.7992655 0.1427 0.013955 0.1153382/0.1700617 3970 1nMVA InLagEVA 0.3020 yo + y, In Lagged EVA,,

150

TABLE 5.11 (continued) STATISTICAL RESULTS OF STEP ONE OF HYPOTHESIS (3) TESTING Panel D nMVA,-I= Model: Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTSR) p-value: Constant IndependentVariable (InTSR) 0.000 0.000 6.47 10.53 0.1249479 0.0193188 0.0870726/0.1628232 0.1202252 0.0114126 0.0978502/0.1426001 4966 1nMVA InTSR 0.0570 + 9, In TSR, o5o

151

TABLE 5.11 (continued) STATISTICAL RESULTS OF STEP ONE OF HYPOTHESIS (3) TESTING Panel E lIn MVA,, Model: Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InTobin'sQ) 0.000 0.000 37.63 39.63 2.8353 0.0753512 2.687584/2.983016 0.4931501 0.0124448 0.4687538/0.5175465 6715 lnMVA InTobin'sQ 0.3480 =
/Tl /To+

In Tobin's Q,,

152

TABLE 5.12 STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel A
In AfVAj, =, 80 +Aln(profitability Model spreadj +A In(Tobin's Q,,) + e,,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant independentVariable (InSPREAD) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InSPREAD) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 33.27 14.01 15.06 3.41377 0.1026209 3.212555/3.614984 0.1477277 0.0105464 0.1270488/0.1684066 0.2639882 0.0175321 0.2296119/0.2983644 3686 1nMVA InSPREAD,InTobin'sQ 0.4311

153

TABLE 5.12 (continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel B

Model In MVA,, = 80 +, 8, ln(profitability spread,,) +A ln(TSR,, + e,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InSPREAD) IndependentVariable (InTSR) p-value: Constant IndependentVariable (InSPREAD) IndependentVariable (InTSR) 0.000 0.000 0.000 25.86 22.05 8.59 2.959312 0.114416 2.734938/3.183686 0.2284903 0.010361 0.208168/0.2488127 0.0918727 0.0106942 0.0709009 2902 1nMVA InSPREAD, InTSR 0.3794

154

TABLE 5.12 (continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel C lInMVAi, Model Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InSPREAD) IndependentVariable (InTIC) p-value: Constant IndependentVariable (InSPREAD) independent Variable (InTIC) 0.000 0.000 0.000 25.92 24.81 15.73 2.399658 0.0925747 2.218154/2.581162 0.2058962 0.0082986 0.1896257/0.2221667 0.5441606 0.0345928 0.4763372/0.6119839 4429 lnMVA InSPREAD,InTIC 0.4029 6,, +, 8, In(profitability spread,,)+ 821n(TIC,, )+e,, =,

155

STATISTICAL

TABLE 5.12 (continued) RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel D
spreadj +A In(Lagged E VAlt)+e,

IIn WA,, 80 +, 8, In(profitability Model =

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InSPREAD) Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InSPREAD) IndependentVariable (InLagEVA) p-value: Constant IndependentVariable (InSPREAD) independentVariable (InLagEVA) 0.000 0.000 I 0.000 I 29.25 24.57 3.20 3.249292 0.1111003 3.031438/3.467147 0.2548422 0.0103714 0.2345052/0.2751793 0.0448538 0.0140203 0.0173616/0.0723459 3261 1nMVA InSPREAD,InLagEVA 0.4030

156

TABLE 5.12(continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel E [In MYA,, Model 80 +, 8, In(total investedcapitalj +)62In(Tobin's Q,, = + e,,

Statistic
Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTIC) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InTIC) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 27.49 16.67 31.66 2.246136 0.081698 2.085977/2.406294 0.6485643 0.0389131 0.5722803/0.7248484 0.4136267 0.013066 0.3880125/0.439241 6712 1nMVA InTIC, InTobin'sQ 0.3784

157

STATISTICAL

TABLE 5.12 (continued) RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel F

Model In MVA,, = 80 + P, In(total investedcapital, +A ln(TSR,, + e,,

Statistic Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTIC) IndependentVariable (InTSR) p-value: Constant independent Variable (InTIC) IndependentVariable (InTSR) 0.000 0.000 0.066 -1.84 19.95 -1.84 -0.0371087 0.020153 -0.0766195/0.0024021 0.8733576 0.0437758 0.7875333 0.1153032 0.0109023 0.0939288/0.1366775 4964 lnMVA InTIC, InTSR 0.1595

158

TABLE 5.12 (continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel G IIn MVAj, 60 +, 61In(Lagged EVA,, A In(Tobin's Model )+ Q,,) + e,, =, Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InLagEVA) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InLagEVA) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 34.33 5.05 30.45 3.369323 0.0981309 3.176901/3.561745 0.06623305 0.0131242 0.040957/0.0919653 0.4924473 0.0161722 0.4607358/0.5241589 3313 lnMVA InLagEVA, InTobin'sQ 0.4277

159

TABLE 5.12 (continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel H IIn WA,, 80 +,Bl In(Lagged EVAJ +A ln(TSR,, Model = ) + e,, Number of Observations DependentVariable IndependentVariable R-Squared. Coefficients: Constant Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InLagEVA) IndependentVariable (InTSR) p-value: Constant IndependentVariable (InLagEVA) IndependentVariable (InTSR) 0.000 0.000 100 14.58 7.63 9.32 0.9106314 0.0624719 0.7881103/1.033152 0.1326416 0.0173883 0.0985394/0.1667438 0.1196001 0.0128377 0.0944226/0.1447776 2576 1nMVA InLagEVA, InTSR 0.3156

160

TABLE 5.12 (continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel I Model In MVAj, =, 80 +Aln(total Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTIC) Std. error 95% C.I. IndependentVariable (InLagEVA) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTIC) IndependentVariable (InLagEVA) p-value: Constant IndependentVariable (InTIC) IndependentVariable (InLagEVA) 0.000 0.000 0.000 9.73 18.01 8.54 0.4753706 0.0488642 0.379562/0.5711792 0.6940391 0.0385467 0.6184601/0.769618 0.1156793 0.0135523 0.0891071/0.1422515 investedcapital,,) 3966 lnMVA InTIC, InLagEVA 0.2844
+ 82 .

In(Lagged EVAJ + e,,

161

TABLE 5.12(continued) STATISTICAL RESULTS OF STEP TWO OF HYPOTHESIS (3) TESTING Panel J IIn MVA,, 8, Model =, +Aln(TSR,, ) +P2 In(Tobin's Qi,) + e, Number of Observations DependentVariable IndependentVariable R-Squared Coefficients: Constant Std. error 95% C.I. IndependentVariable (InTSR) Std. error 95% C.I. IndependentVariable (InTobin'sQ) Std. error 95% C.I. t-statistics: Constant IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) p-value: Constant IndependentVariable (InTSR) IndependentVariable (InTobin'sQ) 0.000 0.000 0.000 29.29 8.55 28.16 2.835656 0.0968174 2.645834/3.025478 0.0940835 0.0110063 0.0725042/0.1156627 0.4528348 0.0160809 0.4213063/0.4843633 4410 1nMVA InTSR, InTobin'sQ 0.3566

162

TABLE 5.13 HYPOTHESIS (3) RELATIVE VALUE-RELEVANCE Panel A Relative Value-RelevanceTest (individual) Profitability Spread IF 38.65% > Tobin's Q Lagged EVA Total Total Invested Shareholder Capital Return - ,> 14.09% 5.70% TEST

34.80% > 30.20% >

Panel B RELATIVE VALUE-RFLEVANCE TEST (pair-wise combinations in order of decreasingpower)

Pair-Wise Combination
Profitability Spread/ Tobin's Q Lagged EVA / Tobin's Q Profitability Spread/ LaggedEVA Profitability Spread/ Total InvestedCapital Return Profitability Spread/ Total Shareholder Total InvestedCapital / Tobin's Q Total ShareholderReturn/ Tobin's Q Return EVA / Total Shareholder Lagged Total InvestedCapital / LaggedEVA

R2
43.11 42.77 40.30 40.29 37.94 37.84 35.66 31.56 28.44

Return 15.95 / Total Shareholder Total Invested Capital

163

TABLE 5.14 HYPOTHESIS 3 INCREMENTAL VALUE-RELEVANCE TEST

Pair-Wise Combination

Incremental Valuerelevance
8.31 7.97 10.10 26.20 32.24 3.04 0.86 25.86 -1.76 10.25 4.46 12.57 1.65 1.64 -0.81 23.75 29.96 1.36 14.35 1.86

Profitability Spread/ Tobin's Q Lagged EVA / Tobin's Q Profitability Spread/ LaggedEVA Profitability Spread/ Total InvestedCapital Profitability Spread/ Total Shareholder Return Total Invested Capital / Tobin's Q Total ShareholderReturn/ Tobin's Q Lagged EVA / Total Shareholder Return Total Invested Capital / LaggedEVA Total Invested Capital / Total Shareholder Return Tobin's Q/ Profitability Spread Tobin's Q/ LaggedEVA Lagged EVA / Profitability Spread Total Invested Capital / Profitability Spread Total ShareholderReturn / Profitability Spread Tobin's Q/ Total InvestedCapital Return Tobin's Q/ Total Shareholder Total ShareholderReturn / LaggedEVA Lagged EVA / Total InvestedCapital ShareholderReturn / Total InvestedCapital Total

164

Appendix C: The 2002 Stern Stewart Performance 1000

Top 1000Creatorsof Shareholderwealth amongU. s. Companies,1998-2002 Market value Added (MVA)

The2002Stern Stewart Performance 1000 Top 1000Creators of Shareholder Wealth AmongU.S. Companies, 1998-2002 MVA
Rank 2002 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 2001 2 3 1 5 12 10 11 14 8 9 6 20 4 21 24 25 19 23 18 16 17 15 13 29 7 30 22 71 27 36 39 1998 Ticker 1 MSFT 6 WIVIT 2 GE 16 JNJ 15 PFE 10 MIRK 20 PG 13 IBM 11 XOM 8 KO 3 INTC 18 DELL C 5 CSCO 51 ORCL 58 AMGN 25 LLY UPS FNM 44 PEP 21 MO 30 ABT 23 HD 62 MMM AIG 48 MDT 28 WYE 74 CMCSA 59 BUD 33 DD 75 PHA Company name Microsoft Corp Wal-Mart Stores General Electric Co Johnson & Johnson Pfizer Inc Merck & Co Procter & Gamble Co Intl Business Machines Corp Exxon Mobil Corp Coca-Cola Co Intel Corp Dell Computer Corp Citigroup Inc Cisco Systems Inc Oracle Corp Amgen Inc Lilly (Eli) & Co United Parcel Service Inc Fannie Mae Pepsico Inc Altria Group Inc Abbott Laboratories Home Depot Inc 3M Co American International Group Medtronic Inc Wyeth Comcast Corp Anheuser-Busch Cos Inc Du Pont (E 1)De Nemours Pharmacia Corp Industry Code 4510 2550 2010 3520 3520 3520 3030 4520 1010 3020 4520 4520 4020 4520 4510 3520 3520 2030 4020 3020 3020 3520 2550 2010 4030 3510 3520 2540 3020 1510 3520 Industry Name Software & Services Retailing Capital Goods Pharmaceuticals& Biotech Pharmaceuticals& Biotech Pharmaceuticals& Biotech Household& Personal Prods Technology Hardware& Equip Energy Food Beverage & Tobacco Technology Hardware& Equip Technology Hardware& Equip Diversified Financials Technology Hardware& Equip Software & Services Pharmaceuticals& Biotech Pharmaceuticals& Biotech Transportation Diversified Financials Food Beverage & Tobacco Food Beverage & Tobacco Pharmaceuticals& Biotech Retailing Capital Goods Insurance Health Care Equipment& Svcs Pharmaceuticals& Biotech Media Food Beverage & Tobacco Materials Pharmaceuticals& Biotech

165

32 33 34 35 36 37 38 39
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

28 35 92 34 49 46 48 33
47 26 37 63 40 59 68 96 53 67 42 54 60 41 55 102 73 85 32 56 31 80 82 45 180 124 38 64 99 97 166 103 128 341 270 51 81 118 111

17 BMY FIRE 63 EBAY AXP 67 CL 98 LOW 198 UNH 73 WAG

WFC KFT BLS BAC TGT FDC QCOM FRX G SYY VZ MIVIC UTX SGP KSS SLM COX DISH MWD FITB TXN GCI EMIR ADP L AFL AMAT LIVIT SO GIS AMZN AVP SYK NXTL BSX GS HDI DNA WWY

Bristol Myers Squibb Federal Home Loan Mortg Corp eBay Inc American Express Colgate-PalmoliveCo Lowes Cos UnitedhealthGroup Inc Walgreen Co
Wells Fargo & Co Kraft Foods Inc Bellsouth Corp Bank Of America Corp Target Corp First Data Corp Qualcomm Inc Forest Laboratories A -CI Gillette Co Sysco Corp Verizon Communications Marsh & Mclennan Cos United Technologies Corp Schering-Plough KohIs Corp SLM Corp Cox Communications -Cl A Echostar Commun Corp -CIA Morgan Stanley Fifth Third Bancorp Texas Instruments Inc Gannett Co Emerson Electric Co Automatic Data Processing Liberty Media Corp -Ser A Aflac Inc Applied Materials Inc Lockheed Martin Corp Southern Co General Mills Inc Amazon. Com Inc Avon Products Stryker Corp Nextel Communications Boston Scientific Corp Goldman Sachs Group Inc Harley-Davidson Inc Genentech Inc Wrigley (Wm) Jr Co

32 69 156 90 344 46 167 27 68 31 140 84 227

52 91 77 72 54 87 148 163 131 53 117 224 127 110 170 173 153

3520 4020 2550 4020 3030 2550 3510 3010 4010 3020 5010 4010 2550 2020 4520 3520 3030 3010 5010 4030 2010 3520 2550 4020 2540 2540 4020 4010 4520 2540 2010 2020 2540 4030 4520 2010 5510 3020 2550 3030 3510 5010 3510 4020 2510 3520 3020

Pharmaceuticals & Biotech Diversified Financials Retailing Diversified Financials Household & Personal Prods Retailing Health Care Equipment & Svcs Food & Drug Retailing Banks Food Beverage & Tobacco Telecommunication Services Banks Retailing Commercial Svcs & Supplies Technology Hardware & Equip Pharmaceuticals & Biotech Household & Personal Prods Food & Drug Retailing Telecommunication Services Insurance Capital Goods Pharmaceuticals & Biotech Retailing Diversified Financials Media Media Diversified Financials Banks Technology Hardware & Equip Media Capital Goods Commercial Svcs & Supplies Media Insurance Technology Hardware & Equip Capital Goods Utilities Food Beverage & Tobacco Retailing Household & Personal Prods Health Care Equipment & Svcs Telecommunication Services Health Care Equipment & Svcs Diversified Financials Automobiles & Components Pharmaceuticals & Biotech Food Beverage & Tobacco

166

79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 ill 112 113 114 115 116 117 118 119 120 121 122
123 124 125

65 75 131 116 100 93 151 89 135 90 58 110 126 105 145 62 109 186 106 179 84 107 50 115 981 222 119 182 181 157 121 138 69 52 146 57 86 156 70 43 108 213 123 176
112 72 136

78 162 47 241 116 179 82 94 187 100 119 36 158 43 243 438 144 125 157 38 145 104 528 151 186 139 101 367 96 176 201 147 264 83 108 109 711 113 246
103 ill .

VIA.B MXIM GPS BBBY K PAYX DOW ITW TRB COST KR13 FOX EMC LLTC PGR DIS ADI APOL LUV ZMH OMC CAT MCD TJX CCU GILD MHP LXK FDX CVC D GDT GD BK SBUX MER BBY EXC KMB BAX SLE STJ GMH AGN
HNZ CAH ALL

Viacom Inc B -Cl Maxim IntegratedProducts Gap Inc Bed Bath & Beyond Inc Kellogg Co PaychexInc Dow Chemical Illinois Tool Works Tribune Co Costco Wholesale Corp Mbna Corp Fox EntertainmentGroup Inc EMC Corp/Ma Linear TechnologyCorp ProgressiveCorp-Ohio Disney (Walt) Co Analog Devices Apollo Group Inc -CIA SouthwestAirlines Zimmer HIdgs Inc Omnicom Group Caterpillar Inc McDonaldsCorp Tjx CompaniesInc Clear Channel Communications Gilead Sciences Inc McGraw-HillCompanies Lexmark Intl Inc -CIA Fedex Corp CablevisionSys Corp -Cl A Dominion ResourcesInc Guidant Corp General DynamicsCorp Bank Of New York Co Inc Starbucks Corp Merrill Lynch & Co Best Buy Co Inc Exelon Corp Kimberly-ClarkCorp Baxter InternationalInc Sara Lee Corp St Jude Medical Inc General Motors Cl H Allergan Inc
Heinz (H J) Co Cardinal Health Inc Allstate Corp

2540 4520 2550 2550 3020 2020 1510 2010 2540 2550 4020 2540 4520 4520 4030 2540 4520 2020 2030 3510 2540 2010 2530 2550 2540 3520 2540 4520 2030 2540 5510 3510 2010 4010 2530 4020 2550 5510 3030 3510 3020 3510 2540 3520 3020 3510 4030

Media Technology Hardware & Equip Retailing Retailing Food Beverage & Tobacco Commercial Svcs & Supplies Materials Capital Goods Media Retailing Diversified Financials Media Technology Hardware & Equip Technology Hardware & Equip Insurance Media Technology Hardware & Equip Commercial Svcs & Supplies Transportation Health Care Equipment & Svcs Media Capital Goods Hotels Restaurants & Leisure Retailing Media Pharmaceuticals & Biotech Media Technology Hardware & Equip Transportation Media Utilities Health Care Equipment & Svcs Capital Goods Banks Hotels Restaurants & Leisure Diversified Financials Retailing Utilities Household & Personal Prods Health Care Equipment & Svcs Food Beverage & Tobacco Health Care Equipment & Svcs Media Pharmaceuticals & Biotech Food Beverage & Tobacco Health Care Equipment & Svcs Insurance

167

126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172

130 122 152 129 104 209 148 173 147 127 114 153 139 155 259 77 185 169 165 178 189 211 174 91 201 78 133 237 191 200 95 192 243 245 140 143 214 184 187 261 216 221 226 132 74 287 175

311 95 317 88 192 222 213 182 189 106 256 343 393 34 271 301 115 305 245 526 183 120 99 136 431 262 292 249 259 135 229 546 188 132 236 293 355 401 270 762 194

WLP CPB AZO PBI XLNX PX HSY CAG DHR NKE STI ITT ADBE ERTS EXPE HPQ BMET MEDI SPLS UCOMA NYT IGT BDX STT CLX AA YUM GDW GENZ INTU MEL HR13 ECL WPO LTD UVN CHIR APID EL WTW` FPL FDO SSP KLAC HCA SYMC ALTR

Wellpoint Hlth Netwrk -CIA Campbell Soup Co Autozone Inc Pitney Bowes Inc Xilinx Inc Praxair Inc Hershey Foods Corp Conagra Foods Inc Danaher Corp Nike Inc -Cl B Suntrust Banks Inc ITT IndustriesInc Adobe Systems Inc. ElectronicArts Inc Expedia Inc Hewlett-PackardCo Biomet Inc MedimmuneInc Staples Inc Unitedglobalcorn -Cl A Inc New York Times Co -Cl A Intl Game Technology Becton Dickinson& Co State Street Corp Clorox Co/De Alcoa Inc Yum Brands Inc Golden West FinancialCorp GenzymeCorp Intuit Inc Mellon FinancialCorp Block H&R Inc Ecolab Inc WashingtonPost -Cl B Limited Brands Inc UnivisionCommunicationsInc Chiron Corp Air Products& ChemicalsInc Lauder Estee Cos Inc -Cl A Weight Watchers Intl Inc FPL Group Inc Family Dollar Stores EW Scripps -CIA Kla-TencorCorp HCA Inc SymantecCorp Altera Corp

3510 3020 2550 2020 4520 1510 3020 3020 2010 2520 4010 2010 4510 4510 4510 4520 3510 3520 2550 2540 2540 2530 3510 4020 3030 1510 2530 4010 3520 4510 4010 2020 1510 2540 2550 2540 3520 1510 3030 2550 5510 2550 2540 4520 3510 4510 4520

HealthCare Equipment Svcs & Food Beverage& Tobacco Retailing CommercialSvcs& Supplies TechnologyHardware& Equip Materials Food Beverage& Tobacco Food Beverage& Tobacco CapitalGoods ConsumerDurables& Apparel Banks CapitalGoods Software& Services Software& Services Software& Services TechnologyHardware& Equip HealthCare Equipment Svcs & Pharmaceuticals Biotech & Retailing Media Media Hotels Restaurants Leisure & HealthCare Equipment Svcs & DiversifiedFinancials Household PersonalProds & Materials HotelsRestaurants Leisure & Banks Pharmaceuticals Biotech & Software& Services Banks Svcs & Supplies Commercial Materials Media Retailing Media & Pharmaceuticals Biotech Materials Household& PersonalProds Retailing Utilities Retailing Media TechnologyHardware& Equip HealthCare Equipment& Svcs Software& Services TechnologyHardware& Equip

168

173 174 175 176 177 178 179

180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219

177 188 266 158 203 247 250

171 134 168 149 162 960 227 232 256 284 142 263 280 241 172 240 225 161 101 76 390 231 255 258 113 450 492 150 224 343 262 276 230 352 234 204 274 217 44 353 .

493 255 514 166 124 184 586

477 302 225 288 37

ACS LIST APA MAR AT PPG BJS

LEH NEM VRTS BHI APC YHOO BEN ATH KRI BF. B IDPH SOTR PPL PEG CTAS DE AVY DGX Cl CE COH BGEN CINF ESRX NTRS MYL CFC MAS RMK EXPD DG ROH DOV VAR NTAP RX ASD PGN CVX BRL

Affiliated COMPSvcs -CIA UST Inc Apache Corp Marriott Intl Inc Alltel Corp PPG Industries Inc BJ Services Co
Lehman Brothers Holdings Inc Newmont Mining Corp Veritas Software Co Baker-Hughes Inc Anadarko Petroleum Corp Yahoolnc Franklin Resources Inc Anthem Inc Knight-Ridder Inc Brown-Forman -Cl B IDEC Pharmaceuticals Corp Southtrust Corp PPL Corp Public Service Entrp Cintas Corp Deere & Co Avery Dennison Corp Quest Diagnostics Inc Cigna Corp Concord Efs Inc CoachInc Biogen Inc Cincinnati Financial Corp Express Scripts Inc Northern Trust Corp Mylan Laboratories Countrywide Financial Corp Masco Corp Aramark Corp Expeditors Intl Wash Inc Dollar General Corp Rohm & Haas Co Dover Corp Varian Medical Systems Inc Network Appliance Inc IMS Health Inc American Standard COSInc Progress Energy Inc Chevrontexaco Corp Barr Laboratories Inc

350 286 575 374 287 207 436 204 986 168 266 193 394 441 196 494 212 214 218 956 276 141 373 327 39 717

4510 3020 1010 2530 5010 1510 1010 4020 1510 4510 1010 1010 4510 4020 3510 2540 3020 3520 4010 5510 5510 2020 2010 2020 3510 3510 2020 2520 3520 4030 3510 4010 3520 4020 2010 2020 2030 2550 1510 2010 3510 4520 3510 2010 5510 1010 3520

Software & Services Food Beverage & Tobacco Energy Hotels Restaurants & Leisure Telecommunication Services Materials Energy Diversified Financials Materials Software & Services Energy Energy Software & Services Diversified Financials Health Care Equipment & Svcs Media Food Beverage & Tobacco Pharmaceuticals & Biotech Banks Utilities Utilities Commercial Svcs & Supplies Capital Goods Commercial Svcs & Supplies Health Care Equipment & Svcs Health Care Equipment & Svcs Commercial Svcs & Supplies Consumer Durables & Apparel Pharmaceuticals & Biotech Insurance Health Care Equipment & Svcs Banks Pharmaceuticals & Biotech Diversified Financials Capital Goods Commercial Svcs & Supplies Transportation Retailing Materials Capital Goods Health Care Equipment & Svcs Technology Hardware & Equip Health Care Equipment & Svcs Capital Goods Utilities Energy Pharmaceuticals & Biotech

169

220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266

326 275 295 183 206 298 272 271 196 220 316 125 325 400 344 301 318 257 292 330 252 332 328 233 317 388 459 198 120 194 260 361 190 279 304 202 83 538 336 160 269 404 223 379 265 299 447

469 319 308 840 306 267 484 640 150 550 758 425 459 281 412 453 269 205 381 601 562 427 230 93 64 404 360 329 388

365 322 210 407 573 238 437 675 806

FTN EOG GPC FISV SNV KMI TSS JCI BBT MCHP BEAS COP PCAR WFMI MTB OXY MKC HLT NFB SIAL CDWC ETN UNP TIF WON CHRW PIXR VVY PCs WMI JP AEE AEP HMA WIVI SDS SCH HCBK FCX USAI RSH FE ABC ANF HET CPS XTO

First Tennessee Natl Corp EOG Resources Inc Genuine Parts Co Fiserv Inc Synovus Financial Cp Kinder Morgan Inc Total System ServicesInc Johnson Controls Inc BB&T Corp Microchip TechnologyInc Bea Systems Inc Conocophillips Paccar Inc Whole Foods Market Inc M&T Bank Corp Occidental PetroleumCorp McCormick & Co Hilton Hotels Corp North Fork Bancorporation Sigma-Aldrich Cdw ComputerCenters Inc Eaton Corp Union Pacific Corp Tiffany & Co Westwood One Inc CH RobinsonWorldwideInc Pixar WeyerhaeuserCo Sprint Pcs Group Waste ManagementInc Jefferson-PilotCorp Ameren Corp American Electric Power Health ManagementAssoc WashingtonMutual Inc Sungard Data SystemsInc Schwab (Charles)Corp Hudson City Bancorp Freeprt Mcmor Cop&GId -Cl B USA Interactive RadioshackCorp FirstenergyCorp AmerisourcebergenCorp Abercrombie& Fitch -Cl A Harrahs EntertainmentInc ChoicepointInc XTO Energy Inc

4010 1010 2550 2020 4010 5510 2020 2510 4010 4520 4510 1010 2510 3010 4010 1010 3020 2530 4010 1510 2550 2010 2030 2550 2540 2030 2540 1510 5010 2020 4030 5510 5510 3510 4010 4510 4020 4010 1510 2550 2550 5510 3510 2550 2530 2020 1010

Banks Energy Retailing Commercial Svcs & Supplies Banks Utilities Commercial Svcs & Supplies Automobiles & Components Banks Technology Hardware & Equip Software & Services Energy Automobiles & Components Food & Drug Retailing Banks Energy Food Beverage & Tobacco Hotels Restaurants & Leisure Banks Materials Retailing Capital Goods Transportation Retailing Media Transportation Media Materials Telecommunication Services Commercial Svcs & Supplies Insurance Utilities Utilities Health Care Equipment & Svcs Banks Software & Services Diversified Financials Banks Materials Retailing Retailing Utilities Health Care Equipment & Svcs Retailing Hotels Restaurants & Leisure Commercial Svcs & Supplies Energy

170

267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299
300 301 302 303 304 305 306 307 308 309 310 311 312 313

322 565 307 293 239 363 238 244 410 338 285 207 342 337 335 541 288 470 461 167 355 368 371 814 302 401 394 378 451 297 387 384 438
340 417 315 423 331 449 366 357 437 516 370 359 264 248

498 WSM 832 DRYR 250 EFX Fil SEIC 203 DNB 295 MOLX 199 UCL ROOM 576 MERQ 253 ED 415 PBG 524 LNCR COL 445 ROST 282 CTXS 303 DJ 756 IRM 870 CECO PNC 280 JBL 456 CIN 540 QLGC 958 WFR TROW JNPR 564 LIZ 510 EAT 617 BLL 332 GWW 790 NVR 911 ETR 972 DF
468 959 495 313 458 134 592 372 613 316 512 348 FAST PFG CMX SRE WAT BCR CCE PDCO LM NWAC XRAY SHW LLL HOT

Williams-SonomaInc Dreyer's Grand Ice Cream Inc Equifax Inc Federated Investors Inc Sei InvestmentsCo Dun & BradstreetCorp Molex Inc Unocal Corp Hotels.Com Mercury InteractiveCorp ConsolidatedEdison Inc Pepsi Bottling Group Inc Lincare Holdings Inc Rockwell Collins Inc Ross Stores Inc Citrix Systems Inc Dow Jones & Co Inc Iron Mountain Inc Career EducationCorp PNC FinancialSvcs Group Inc Jabil Circuit Inc Cinergy Corp Qlogic Corp Memc ElectronicMatrialsInc Price (T. Rowe) Group Juniper NetworksInc Liz Claiborne Inc Brinker Intl Inc Ball Corp Grainger (W W) Inc NVR Inc Entergy Corp Dean Foods Co
Fastenal Co Principal Financial Grp Inc Caremark Rx Inc Sempra Energy Waters Corp Bard (C. R.) Inc Coca-Cola Enterprises Patterson Dental Co Legg Masoninc Northwest Airlines Corp Dentsply Internatl Inc Sherwin-Williams Co L-3 Communications Hldgs Inc Starwood Hotels&Resorts Wrld

2550 3020 2020 4020 4020 2020 4520 1010 4510 4510 5510 3020 3510 2010 2550 4510 2540 2020 2020 4010 4520 5510 4520 4520 4020 4520 2520 2530 1510 2010 2520 5510 3020 2010 4020 3510 5510 4520 3510 3020 3510 4020 2030 3510 2550 2010 2530

Retailing Food Beverage & Tobacco Commercial Svcs & Supplies Diversified Financials Diversified Financials Commercial Svcs & Supplies Technology Hardware & Equip Energy Software & Services Software & Services Utilities Food Beverage & Tobacco Health Care Equipment & Svcs Capital Goods Retailing Software & Services Media Commercial Svcs & Supplies Commercial Svcs & Supplies Banks Technology Hardware & Equip Utilities Technology Hardware & Equip Technology Hardware & Equip Diversified Financials Technology Hardware & Equip Consumer Durables & Apparel Hotels Restaurants & Leisure Materials Capital Goods Consumer Durables & Apparel Utilities Food Beverage & Tobacco Capital Goods Diversified Financials Health Care Equipment & Svcs Utilities Technology Hardware & Equip Health Care Equipment & Svcs Food Beverage & Tobacco Health Care Equipment & Svcs Diversified Financials Transportation Health Care Equipment & Svcs Retailing Capital Goods Hotels Restaurants & Leisure

171

314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360

212 372 381 554 345 154 402 432 413 267 334 393 306 350 442 365 235 455 425 982 380 309 506 303 532 314 436 376 291 570 375 544 319 673 462 254 563 61 141 435 430 426 420 278 674 700 312

979 LH 623 ACV 448 DLX PTV 687 FHCC MET 334 IFF 937 OEI 248 ABI 467 RHI 352 HB 451 HRL ABK TCB 896 PTEN 718 ADVP 391 PSFT CBSS CBH 974 3UALAQ 429 DTE 336 LEG BPOP 509 ESV 867 CHS 312 PH 616 S11 NCF 421 DLTR 872 COCO 635 MUR NYB 290 VMC 338 RCNC 261 CTL SPC 347 SEPR 122 DUK 178 CD 385 OSI UB 560 NDN 416 GNTX 447 OHP DRL 917 AMLN MI

Laboratory Cp Of Amer HIdgs Alberto-Culver Co B -Cl Deluxe Corp Pactiv Corp First Health Group Corp Metlife Inc Intl Flavors & Fragrances Ocean Energy Inc Applera Corp Applied Biosys Robert Half Intl Inc Hillenbrand Industries Hormel Foods Corp Ambac Financial Gp TCF Financial Corp Patterson-UtiEnergy Inc Advancepcs Peoplesoft Inc Compass BancsharesInc Commerce Bancorp Inc/Nj UAL Corp Dte Energy Co Leggett & Platt Inc Popular Inc Ensco InternationalInc Chicos Fas Inc Parker-HannifinCorp Smith InternationalInc National CommerceFinancial Dollar Tree Stores Inc CorinthianColleges Inc Murphy Oil Corp New York Cmnty BancorpInc Vulcan MaterialsCo RCN Corp Centurytel Inc St Paul Cos Sepracor Inc Duke Energy Corp Cendant Corp Outback SteakhouseInc UnionbancalCorp 99 Cents Only Stores Gentex Corp Oxford Health Plans Inc Doral Financial Corp Amylin PharmaceuticalsInc Marshall & Ilsley Corp

3510 3030 2020 1510 3510 4030 1510 1010 3510 2020 3510 3020 4030 4010 1010 3510 4510 4010 4010 2030 5510 2520 4010 1010 2550 2010 1010 4010 2550 2020 1010 4010 1510 5010 5010 4030 3520 5510 2020 2530 4010 2550 2510 3510 4020 3520 4010

Health Care Equipment & Svcs Household & Personal Prods Commercial Svcs & Supplies Materials Health Care Equipment & Svcs Insurance Materials Energy Health Care Equipment & Svcs Commercial Svcs & Supplies Health Care Equipment & Svcs Food Beverage & Tobacco Insurance Banks Energy Health Care Equipment & Svcs Software & Services Banks Banks Transportation Utilities Consumer Durables & Apparel Banks Energy Retailing Capital Goods Energy Banks Retailing Commercial Svcs & Supplies Energy Banks Materials Telecommunication Services Telecommunication Services Insurance Pharmaceuticals & Biotech Utilities Commercial Svcs & Supplies Hotels Restaurants & Leisure Banks Retailing Automobiles & Components Health Care Equipment & Svcs Diversified Financials Pharmaceuticals & Biotech Banks

172

361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379
380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407

489 483 385 837 477 503 159 360 472 286 445 229 704 431 452 382 832 611 579
457 480 766 268 646 429 453 484 491 323 349 518 661 399 631 556 501 416 508 310 583 369 374 555 163 586 428 624

254 57 608 854

678 432 298 521 798 531 507 476 929 383
655 486 961 424 733 422 650 652

BRO ASO SEE LVLT DBD EQT A AJG JEC DRI VFC NVLS ENDP ETIVI FSH LAMR JBLU CVH APCC
MNI MDP KMX DST RCII KKD HSP APH SCG TRH NEU HAR CELG FLR SCIO TBL GRA UTSI EDIVIC DVN PPP EC CEY ISCA USB COLM BMS FIC

Brown & Brown Inc Amsouth Bancorporation Sealed Air Corp Level 3 Commun Inc Diebold Inc Equitable ResourcesInc Agilent TechnologiesInc Gallagher (Arthur J.) & Co Jacobs EngineeringGroup Inc Darden RestaurantsInc Vf Corp Novellus Systems Inc Endo PharmaceuticalsHldgs Entercom Communications Corp Fisher Scientific Intl Inc Lamar Advertising Co -CIA Jetblue Airways Corp Coventry Health Care American Pwr Cnvrsion
Mcclatchy Co -CIA Meredith Corp Carmax Inc Dst Systems Inc Rent-A-Center Inc Krispy Kreme Doughnuts Inc Hispanic Broadcasting -Cl A Amphenol Corp Scana Corp Transatlantic Holdings Inc Neuberger Berman Inc Harman International Inds Celgene Corp Fluor Corp Scios Inc Timberland Co -CIA Grace (W R) & Co Utstarcom Inc Education Management Corp Devon Energy Corp Pogo Producing Co Engelhard Corp Certegy Inc Intl Speedway Corp -CIA US Bancorp Columbia Sportswear Co Bemis Co Fair Isaac Corp

861 809 644 855 738 395 728 559 716 499 435 866 538 815

4030 4010 1510 5010 4520 5510 4520 4030 2010 2530 2520 4520 3520 2540 3510 2540 2030 3510 2010 2540 2540 2550 2020 2550 2530 2540 4520 5510 4030 4020 2520 3520 2010 3520 2520 1510 4520 2020 1010 1010 1510 2020 2530 4010 2520 1510 4510

Insurance Banks Materials Telecommunication Services Technology Hardware & Equip Utilities Technology Hardware & Equip Insurance Capital Goods Hotels Restaurants & Leisure Consumer Durables & Apparel Technology Hardware & Equip Pharmaceuticals & Biotech Media Health Care Equipment & Svcs Media Transportation Health Care Equipment & Svcs Capital Goods Media Media Retailing Commercial Svcs & Supplies Retailing Hotels Restaurants & Leisure Media Technology Hardware & Equip Utilities Insurance Diversified Financials Consumer Durables & Apparel Pharmaceuticals & Biotech Capital Goods Pharmaceuticals & Biotech Consumer Durables & Apparel Materials Technology Hardware & Equip Commercial Svcs & Supplies Energy Energy Materials Commercial Svcs & Supplies Hotels Restaurants & Leisure Banks Consumer Durables & Apparel Materials Software & Services

173

408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 . 444 445 446 447 448 449 450 451 452 453 454

524 958 485 713 197 581 389 638 564 515 471 475 347 246 785 412 478 414 649 362 406 539 548 521 281 572 535 373 205 750 421 210 591 486 218 481 607 527 561 664 500 575 513 495 619 576

296 RSG 181 AES EV CZN JHF 705 ZBRA TMK 274 MAT FVB 901 SCRI ERIE 551 VAL 845 CEPH AOC 940 IVX 654 LEN 645 DCI 748 ATK NXTP 430 WEN 604 BEC 536 PLL 794 EXBD BNK RGC 452 MGG 704 STR 772 GTK 519 FO CB 922 WDC 769 CAKE 155 MAY ENR 523 CNX 208 TSG 462 DV 535 REV 690 PAX 362 DL MKL VLY 732 ESI PKG UDI 888 SRCL 891 ICST

Republic Services Inc AES Corp. (The) Eaton Vance Corp Citizens CommunicationsCo Hancock John Finl Svcs Inc Zebra TechnologiesCp A -Cl Torchmark Corp Mattel Inc First Virginia Banks Inc Sicor Inc Erie IndemnityCo -CIA Valspar Corp Cephalon Inc Aon Corp Ivax Corp Lennar Corp DonaldsonCo Inc Alliant TechsystemsInc Nextel Partners Inc Wendy'S InternationalInc Beckman Coulter Inc Pall Corp Corporate ExecutiveBrd Co BanknorthGroup Inc Regal EntertainmentGroup Mgm Mirage Questar Corp Gtech HoldingsCorp Fortune Brands Inc Chubb Corp Western Digital Corp CheesecakeFactory Inc May DepartmentStores Co EnergizerHIdgs Inc Consol Energy Inc Sabre HIdgs Corp -CIA Devry Inc Revlon Inc -CIA Paxson Comm Corp -Cl A Dial Corporation Markel Corp Valley National Bancorp Itt EducationalSvcs Inc PackagingCorp Of America United Defense Industries Stericycle Inc IntegratedCircuit Systems

2020 5510 4020 5010 4030 4520 4030 2520 4010 3520 4030 1510 3520 4030 3520 2520 2010 2010 5010 2530 3510 2010 2020 4010 2540 2530 5510 2530 2520 4030 4520 2530 2550 3030 1510 2020 2020 3030 2540 3030 4030 4010 2020 1510 2010 2020 4520

Commercial Svcs & Supplies Utilities Diversified Financials Telecommunication Services Insurance Technology Hardware & Equip Insurance Consumer Durables & Apparel Banks Pharmaceuticals & Biotech Insurance Materials Pharmaceuticals & Biotech Insurance Pharmaceuticals & Biotech Consumer Durables & Apparel Capital Goods Capital Goods Telecommunication Services Hotels Restaurants & Leisure Health Care Equipment & Svcs Capital Goods Commercial Svcs & Supplies Banks Media Hotels Restaurants & Leisure Utilities Hotels Restaurants & Leisure Consumer Durables & Apparel Insurance Technology Hardware & Equip Hotels Restaurants & Leisure Retailing Household & Personal Prods Materials Commercial Svcs & Supplies Commercial Svcs & Supplies Household & Personal Prods Media Household & Personal Prods Insurance Banks Commercial Svcs & Supplies Materials Capital Goods Commercial Svcs & Supplies Technology Hardware & Equip

174

455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501

469 489 351 239 466 779 354 665 559 443 463 202 620 450 473 597 786 460 313 345 626 424 396 574 409 339 219 715 777 783 440 611 651 774 448 569 801 531 547 441 598 446 565 588 789 596 803 329 603 507 722 546 702 545 600 540 818 487 723 729 862 454 836 567 679 290 587 577 578 236 537 144 520 726 670 774 689 629 730 609 913 602

HHS SVM STZ DHI SNPS NSC ICOS NATI CFFN AAP JNY STU VC1 CIF CAL NVDA ADTN MAN BER FNF IFIN MRX JW.A UHS CXR APP13 PETM MHK REY HLR LEE PHM TLB UGI PIR PSC BSG MIL JNS ADS COF HSIC STN HBAN FULT LYO PETC

Harte Hanks Inc ServicemasterCo Constellation Brands A -Cl DR Horton Inc SynopsysInc Norfolk Southern Corp Icos Corp National InstrumentsCorp Capitol Federal Financial Advance Auto Parts Jones Apparel Group Inc Student Loan Corp Valassis CommunicationsInc Charter One Finl Inc ContinentalAirls Inc B -Cl Nvidia Corp Adtran Inc Manpower Inc/Wi Berkley (W R) Corp Fidelity National Finl Inc Investors Financial Svcs Cp Medicis PharmaceutCp -Cl A Wiley (John) & Sons -Cl A Universal Health Svcs -Cl B Cox Radio Inc -CI A Applebees Intl Inc Petsmart Inc Mohawk IndustriesInc Reynolds & Reynolds -Cl A Hollinger Intl Inc -Cl A Lee Enterprises Pulte Homes Inc Talbots Inc Ugi Corp Pier 1 Imports Inc/De PhiladelphiaSuburbanCorp Bisys Group Inc Millipore Corp Janus Capital Group Inc Alliance Data Systems Corp Capital One Fini Corp Schein Henry Inc Station Casinos Inc HuntingtonBancshares Fulton Financial Corp Lyondell ChemicalCo Petco Animal Supplies Inc

2540 2020 3020 2520 4510 2030 3520 4510 4010 2550 2520 4020 2020 4010 2030 4520 4520 2020 4030 4030 4020 3520 2540 3510 2540 2530 2550 2520 2020 2540 2540 2520 2550 5510 2550 5510 2020 3510 4020 2020 4020 3510 2530 4010 4010 1510 2550

Media CommercialSvcs& Supplies Food Beverage& Tobacco ConsumerDurables& Apparel Software& Services Transportation Pharmaceuticals Biotech & Software& Services Banks Retailing ConsumerDurables& Apparel DiversifiedFinancials CommercialSvcs& Supplies Banks Transportation TechnologyHardware& Equip TechnologyHardware& Equip CommercialSvcs& Supplies Insurance Insurance DiversifiedFinancials Pharmaceuticals Biotech & Media HealthCare Equipment Svcs & Media Hotels Restaurants Leisure & Retailing ConsumerDurables& Apparel CommercialSvcs& Supplies Media Media ConsumerDurables& Apparel Retailing Utilities Retailing Utilities CommercialSvcs & Supplies & HealthCare Equipment Svcs DiversifiedFinancials CommercialSvcs & Supplies DiversifiedFinancials & HealthCare Equipment Svcs & Hotels Restaurants Leisure Banks Banks Materials Retailing

175

502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

592 635 434 542 566 552 703 650 676 458 87 728 534 405 816 525 568 736 778 585 468 573 242 781 533 666 512 636 735 550 706 488 403 639 608 377 711 595 748 543 644 764 623 439 562 605 686

541 HNI 643 LANC 899 MIK WL CBSH 681 RCI 780 ATG AMTD 686 CHK MRBK 70 SWY 767 SMG MCY 471 WHR 814 SNDK 823 WSTC 596 ORLY LUK 788 STK 402 HUB.13 483 RL 612 MEG LNC GPT 701 RDC 683 CHD 520 MLHR 727 MDU 984 KSE CYN 590 VHI 633 ALE BE 629 NFG 753 SFD, 333 NUE 321 WWCA 571 BLC 671 PSUN 529 CEG 530 WEC SAFC 856 PFGC 878 HNT 544 NBL 570 CBT 341 BR

Hon Industries Lancaster Colony Corp Michaels Stores Inc Wilmington Trust Corp Commerce BancsharesInc Renal Care Group Inc AgI Resources Inc Ameritrade Holding Corp Chesapeake EnergyCorp Mercantile BanksharesCorp Safeway Inc Scoffs Co Mercury General Corp Whirlpool Corp Sandisk Corp West Corp 0 Reilly Automotive Inc Leucadia NationalCorp Storage TechnologyCp Hubbell Inc -CI B Polo Ralph Lauren Cp -Cl A Media General -CIA Lincoln NationalCorp Greenpoint FinancialCorp Rowan Cos Inc Church & Dwight Inc Miller (Herman) Inc Mdu ResourcesGroup Inc Keyspan Corp City National Corp Valhi Inc Allete Inc BearingpointInc National Fuel Gas Co Smithfield Foods Inc Nucor Corp Western Wireless Corp -Cl A Belo Corp -Ser A Com Pacific SunwearCalif Inc ConstellationEnergyGrp Inc Wisconsin Energy Corp Safeco Corp PerformanceFood Group Co Health Net Inc - Cl A Noble Energy Inc Cabot Corp Burlington ResourcesInc

2020 3020 2550 4010 4010 3510 5510 4020 1010 4010 3010 1510 4030 2520 4520 2020 2550 4020 4520 2010 2520 2540 4030 4010 1010 3030 2020 5510 5510 4010 1510 2010 4510 5510 3020 1510 5010 2540 2550 5510 5510 4030 3010 3510 1010 1510 1010

Commercial Svcs & Supplies Food Beverage & Tobacco Retailing Banks Banks Health Care Equipment & Svcs Utilities Diversified Financials Energy Banks Food & Drug Retailing Materials Insurance Consumer Durables & Apparel Technology Hardware & Equip Commercial Svcs & Supplies Retailing Diversified Financials Technology Hardware & Equip Capital Goods Consumer Durables & Apparel Media Insurance Banks Energy Household & Personal Prods Commercial Svcs & Supplies Utilities Utilities Banks Materials Capital Goods Software & Services Utilities Food Beverage & Tobacco Materials Telecommunication Services Media Retailing Utilities Utilities Insurance Food & Drug Retailing Health Care Equipment & Svcs Energy Materials Energy

176

549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594

755 652 648 773 731 716 610 618 797 744 482 612 621 625 391 498 465 616 530 514 511 609 614 758 690 654 668 324 779 763 547 427 681 493 557 687 683 770 918 630 633 672 283 289 662 720

751 WERN 737 GYI 787 CBRL 793 J1BHT 668 GXP EW CRL 606 OCR 240 ROK 810 ARG 482 SON 618 NFX 619 NST 744 IDC 502 DPL 457 SFA 297 JWN 591 SWFT 638 HLYW 782 RI 850 ELX XJT 664 MTD 567 CSL 653 JDEC FTI 800 VVC 433 CVG BOH MCCC CAM ZION CFR CTX GRP PPDI AME CLE CKFR RYL HSC KEG MTG CDN HCC PNY

736

463 834 764 505 406 898 680 943 481 731

Werner EnterprisesInc Getty Images Inc Cbrl Group Inc Hunt (Jb) Transprt Svcs Inc Great Plains Energy Inc Edwards LifesciencesCorp Charles River Labs Intl Inc Omnicare Inc RockwellAutomation Airgas Inc Sonoco ProductsCo Newfield ExplorationCo, Nstar InteractiveData Corp DpI Inc Scientific-AtlantaInc NordstromInc Swift TransportationCo,Inc HollywoodEntmt Corp Ruby Tuesday Inc Emulex Corp ExpressjetHoldings Inc Mettler-ToledoIntl Inc Carlisle Cos Inc EdwardsJD& Co Fmc TechnologiesInc Vectren Corp ConvergysCorp Bank Of Hawaii Corp MediacomCommunications Corp Cooper Cameron Corp Zions Bancorporation Cullen/FrostBankers Inc Centex Corp Grant Prideco Inc PharmaceuticalProd Dev Inc Ametek Inc Claires Stores Inc Checkfree Corp Ryland Group Inc Harsco Corp Key Energy Services Inc Mgic InvestmentCorp/Wi Cadence Design Sys Inc Hcc Ins HIdgs Inc Natural Gas Co Piedmont

2030 2540 2530 2030 5510 3510 3520 3510 2010 1510 1510 1010 5510 2540 5510 4520 2550 2030 2550 2530 4520 2030 4520 2010 4510 1010 5510 2020 4010 2540 1010 4010 4010 2520 1010 3510 2010 2550 2020 2520 2010 1010 4030 4510 4030 5510

Transportation Media Hotels Restaurants Leisure & Transportation Utilities HealthCare Equipment Svcs & Pharmaceuticals Biotech & HealthCare Equipment Svcs & CapitalGoods Materials Materials Energy Utilities Media Utilities TechnologyHardware& Equip Retailing Transportation Retailing HotelsRestaurants Leisure & TechnologyHardware& Equip Transportation TechnologyHardware& Equip CapitalGoods Software& Services Energy Utilities Svcs& Supplies Commercial Banks Media Energy Banks Banks ConsumerDurables& Apparel Energy HealthCare Equipment& Svcs CapitalGoods Retailing CommercialSvcs & Supplies ConsumerDurables& Apparel CapitalGoods Energy Insurance Software& Services Insurance Utilities

177

595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641

688 597 701 479 658 882 504 476 528 772 600 768 617 718 637 339 738 496 137 788 678 694 714 659 737 294 517 726 708 671 395 835 697 752 396 696 803 888 685 754 771 749 742 817 669 692 474

694 953 390 658 777 835 409 461 925 709 831 871 674 228 696 684 285 129 880 177 697 691 318 323 552 868 934 369 824 712 932 714 863 742 670 799 747 676 830 980 828 85

HIB ADSK WGR AOT LZ WR TOL KMG AEOS HOV KBH SJM VRC RPM ONNN MYG PGL MBG SEBL NWL NBTY BNI ALB AHG CYCL PCG TFX BOKF GREY TEK DO SBGI ATR AGI NCC RGS EGN SIRI BTH STE WGL CYT WPS PXD MDC BSC CA

Hibernia Corp -CIA Autodesk Inc Western Gas ResourcesInc Apogent Technologies Inc Lubrizol Corp Westar Energy Inc Toll Brothers Inc Kerr-McgeeCorp Amern Eagle Outfitters Inc Hovnanian Entrprs Inc A -Cl Kb Home Smucker (Jm) Co Varco InternationalInc Rpm InternationalInc On SemiconductorCorp Maytag Corp Peoples Energy Corp Mandalay Resort Group Siebel Systems Inc Newell RubbermaidInc Nbty Inc BurlingtonNorthern Santa Fe Albemarle Corp Apria HealthcareGroup CentennialCommun Cp -CIA Pg&E Corp Teleflex Inc Bok FinancialCorp Grey Global Group Inc Tektronix Inc DiamondOffshre Drilling Inc Sinclair BroadcastGp -Cl A Aptargroup Inc Alliance Gaming Corp NationalCity Corp Regis Corp/Mn EnergenCorp Sirius Satellite Radio Inc Blyth Inc Steris Corp WgI Holdings Inc Cytec IndustriesInc Wps ResourcesCorp Pioneer Natural ResourcesCo Mdc Holdings Inc Bear Stearns CompaniesInc ComputerAssociates Intl Inc

4010 4510 1010 3510 1510 5510 2520 1010 2550 2520 2520 3020 1010 1510 4520 2520 5510 2530 4510 2520 3030 2030 1510 3510 5010 5510 2010 4010 2540 4520 1010 2540 1510 2530 4010 2550 5510 2540 2520 3510 5510 1510 5510 1010 2520 4020 4510

Banks Software& Services Energy HealthCare Equipment Svcs & Materials Utilities ConsumerDurables& Apparel Energy Retailing ConsumerDurables& Apparel ConsumerDurables& Apparel Food Beverage& Tobacco Energy Materials TechnologyHardware& Equip ConsumerDurables& Apparel Utilities Hotels Restaurants Leisure & Software& Services ConsumerDurables& Apparel Household& PersonalProds Transportation Materials HealthCare Equipment Svcs & Telecommunication Services Utilities CapitalGoods Banks Media TechnologyHardware& Equip Energy Media Materials & Hotels Restaurants Leisure Banks Retailing Utilities Media ConsumerDurables& Apparel HealthCare Equipment& Svcs Utilities Materials Utilities Energy ConsumerDurables& Apparel DiversifiedFinancials Software& Services

178

642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688

767 634 784 759 587 215 632 820 795 747 300 551 613 796 208 840 677 397 745 418 741 792 606 894 809 599 722 815 906 852 560 857 762 775 808 791 598 812 655 721 693 398 780 757 739 383 695

906 TKR 410 NI

PCO

487 513 357 265 894 785 792 649 775 642 626 685

920 660 752 549 384 740 405 661 981 858 768 826 707 933 849 595 914 746 632 820 251 440

SE PNR BRCD SSCC HAS RBK WCN MBI SCHL RKY ATO KG PBCT OGE AMKR PL AGE BYD PSID AXL XMSR BGG BDK SXT SEM PRM FAF CMH OLN CMLS ASBC COKE BKH1 LZB BTU NOI HE ZLC LRCX WOR SNA RYN MRO HTV

Timken Co Nisource Inc Premcor Inc 7-Eleven Inc Pentair Inc Brocade CommunicationsSys Smurfit-StoneContainerCorp Hasbro Inc Reebok InternationalLtd Waste ConnectionsInc Mbia Inc Scholastic Corp Coors (Adolph) B -Cl Atmos Energy Corp King PharmaceuticalsInc Peoples Bank BridgeportCt Oge Energy Corp Amkor TechnologyInc Protective Life Corp Edwards (A G) Inc Boyd Gaming Corp Puget Energy Inc American Axle & Mfg Hldgs Xm Satellite Radio Hldgs Inc Briggs & Stratton Black & Decker Corp SensientTechnologiesCorp Select Medical Corp Primedia Inc First American Corp/Ca Clayton Homes Inc Olin Corp Cumulus Media Inc Associated Banc Corp Coca-Cola BtIng Cons Black Hills Corp La-Z-Boy Inc Peabody Energy Corp National-OilwellInc Hawaiian Electdc Inds Zale Corp Lam ResearchCorp Worthington Industries Snap-On Inc Rayonier Inc Marathon oil Corp Hearst-ArgyleTelevision

2010 5510 1010 3010 2010 4520 1510 2520 2520 2020 4030 2540 3020 5510 3520 4010 5510 4520 4030 4020 2530 5510 2510 2540 2010 2520 3020 3510 2540 4030 2520 1510 2540 4010 3020 5510 2520 1510 1010 5510 2550 4520 1510 2520 1510 1010 2540

Capital Goods Utilities Energy Food & Drug Retailing Capital Goods TechnologyHardware& Equip Materials ConsumerDurables& Apparel ConsumerDurables& Apparel CommercialSvcs & Supplies Insurance Media Food Beverage& Tobacco Utilities Pharmaceuticals Biotech & Banks Utilities TechnologyHardware& Equip Insurance DiversifiedFinancials HotelsRestaurants Leisure & Utilities Automobiles& Components Media CapitalGoods ConsumerDurables& Apparel Food Beverage& Tobacco HealthCare Equipment Svcs & Media Insurance ConsumerDurables& Apparel Materials Media Banks Food Beverage& Tobacco Utilities ConsumerDurables& Apparel Materials Energy Utilities Retailing TechnologyHardware& Equip Materials ConsumerDurables& Apparel Materials Energy Media

179

689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712
713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735

578 675 582 725 364 765 727 843 761 724 499 443 509 679 710 698 641 682 793 422 790 667 789
801 855 743 66 419 707 571 356 753 805 647 594 858 494 699 822 645 549 884 523 367 868 320

455 539 659 545 503 816 897 700 797 375

400 639 945 881 624 725 419 478

358 215 80 918 426 971 743 875 466 284 741 720 534 479

HCR CR GAS CK SWK TPC SLGN ROAD GET NMG.A CYH DNY UTR TVL MLM PSS PDE ROIAK AGY SKYF AVX FOE NAV DQE
BOL WRC KSU EDS IRF ORI EMN NSM EMMS UW

Manor Care Inc Crane Co Nicor Inc Crompton Corp Stanley Works Triton Pcs HIdgs Inc Silgan Holdings Inc Roadway Corp Gaylord Entertainment Neiman-MarcusGroup Inc Community Health SystemsInc Donnelley (R R) & Sons Co Unitrin Inc Lin Tv Corp Martin Marietta Materials Payless ShoesourceInc Pride InternationalInc Radio One Inc Argosy Gaming Corp Sky Financial Group Inc Avx Corp Ferro Corp Navistar Internationl Dqe Inc
Bausch & Lomb Inc Westport Resources Corp Kansas City Southern Electronic Data Systems Corp Intl Rectifier Corp Old Republic Intl Corp Eastman Chemical Co National Semiconductor Corp Emmis Communictns Cp -CIA Universal CorpNa Pepco Holdings Inc Winn-Dixie Stores Inc Borg Warner Inc Radian Group Inc Harris Corp Cnf Inc Linens N Things Inc Pmi Group Inc Regions Finl Corp Readers Digest Assn Bjs Wholesale Club Inc Jo-Ann Stores Inc -Cl A Spx Corp

POM
WIN BWA RDN HRS CNF LIN

PIVII
354 473 919 569 RF RDA BJ JAS. A SPW

3510 2010 5510 1510 2520 5010 1510 2030 2530 2550 3510 2020 4030 2540 1510 2550 1010 2540 2530 4010 4520 1510 2010 5510 3510 1010 2030 4510 4520 4030 1510 4520 2540 3020 5510 3010 2510 4030 4520 2030 2550 4030 4010 2540 2550 2550 2010

Health Care Equipment & Svcs Capital Goods Utilities Materials Consumer Durables & Apparel Telecommunication Services Materials Transportation Hotels Restaurants & Leisure Retailing Health Care Equipment & Svcs Commercial Svcs & Supplies Insurance Media Materials Retailing Energy Media Hotels Restaurants & Leisure Banks Technology Hardware & Equip Materials Capital Goods Utilities Health Care Equipment & Svcs Energy Transportation Software & Services Technology Hardware & Equip Insurance Materials Technology Hardware & Equip Media Food Beverage & Tobacco Utilities Food & Drug Retailing Automobiles & Components Insurance Technology Hardware & Equip Transportation Retailing Insurance Banks Media Retailing Retailing Capital Goods

180

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782

717 807 776 838 819 733 800 628 296 305 914 898 851 760 282 467 846 828 902 830 818 863 799 806 732 841 833 869 834 842 827 867 823 896 590 769 848 660 656 864 734 829 740 847 558 831 892

778 902 669 380 695 829 588 500 525 364 628 754 368 804 955 378 472 885 923 504 837 948 847 964 889 795 965 841 909 938 446 825 821 689 607 882 985 977 844 593 912

GPI SPF ANN SBL CAO CNL BOW FIVIER TDS XEL HPC LII ABF IDA TER ISIL SUG ACI RAD GLK WSC WBS BZH ACXM PCP KMT MXO OKE IPX SWX TMO ATAH ISLE BWS BKS FST AEN LNT IBC ICCI ESA AG DVA PCH ARM MEE PCU

Group 1 Automotive Inc Standard Pacific Cp Anntaylor Stores Corp Symbol Technologies Csk Auto Corp Cleco Corp Bowater Inc Firstmerit Corp Telephone & Data Xcel Energy Inc Hercules Inc Lennox International Inc Airborne Inc Idacorp Inc Teradyne Inc Intersil Corp -CIA Southern Union Co Arch Coal Inc Rite Aid Corp Great Lakes Chemical Corp Wesco Financial Corp Webster Financial Corp Beazer Homes Usa Inc Acxiom Corp Precision Castparts Corp Kennametal Inc Maxtor Corp Oneok Inc Interpool Inc Southwest Gas Corp Thermo Electron Corp Ata Holdings Corp isle Of Capri Casinos Inc Brown Shoe Inc Barnes & Noble Inc Forest Oil Corp Amc Entertainment Inc Alliant Energy Corp Interstate Bakeries Cp Insight Communications Inc Extended Stay America Inc Agco Corp Davita Inc Potlatch Corp Arvinmeritor Inc Massey Energy Co Southern Peru Copper

2550 2520 2550 4520 2550 5510 1510 4010 5010 5510 1510 2010 2030 5510 4520 4520 5510 1510 3010 1510 2020 4010 2520 4510 2010 2010 4520 5510 2020 5510 4520 2030 2530 2520 2550 1010 2540 5510 3020 2540 2530 2010 3510 1510 2510 1510 1510

Retailing Consumer Durables & Apparel Retailing Technology Hardware & Equip Retailing Utilities Materials Banks Telecommunication Services Utilities Materials Capital Goods Transportation Utilities Technology Hardware & Equip Technology Hardware & Equip Utilities Materials Food & Drug Retailing Materials Commercial Svcs & Supplies Banks Consumer Durables & Apparel Software & Services Capital Goods Capital Goods Technology Hardware & Equip Utilities Commercial Svcs & Supplies Utilities Technology Hardware & Equip Transportation Hotels Restaurants & Leisure Consumer Durables & Apparel Retailing Energy Media Utilities Food Beverage & Tobacco Media Hotels Restaurants & Leisure Capital Goods Health Care Equipment & Svcs Materials Automobiles & Components Materials Materials

181

783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829

665 277 746 870 861 519 622 871 709 663 813 927 877 897 715 444 891 811 702 589 878 810 856 798 825 844 845 604 824 584 874 794 883 723 653 890 684 502 802 895 642 849 893 885 860 854 756

TIN CMA 893 VLO 873 R 819 VPI 582 AN 857 TDW 692 PZ13 663 JBX WC1 647 BCC 326 WP1 924 CNJ 577 SUN 883 BGP FCS 161 BMC AF 518 LEA 887 TTN 846 IVGN 389 IM 585 PKI 363 ATI ABG 710 TEX 954 UAG 892 SAH 944 CHRS KEY 990 GHVI 759 BC 886 AVA 666 CTB 634 LAF 808 ALK UPC 839 TECD 557 EAS 966 OSG 852 FLS 876 URS 928 CPO UCO 805 NOR 376 Sol 706 PNW

515

Temple-Inland Inc Comerica Inc. Valero Energy Corp Ryder System Inc Vintage Petroleum Inc Autonation Inc Tidewater Inc Pittston Co, Jack In The Box Inc Wci Communities Inc Boise Cascade Corp Watson Pharmaceuticals Inc Cole National Corp Sunocolnc Borders Group Inc Fairchild Semiconductor Intl Bmc Software Inc Astoria Finl Corp Lear Corp Titan Corp Invitrogen Corp Ingram Micro Inc -Cl A Perkinelmer Inc Allegheny Technologies Inc Asbury Automotive Group Inc Terex Corp United Auto Group Inc Sonic Automotive Inc -CIA Charming Shoppes Keycorp Genesis Health Ventures Inc Brunswick Corp Avista Corp Cooper Tire & Rubber Lafarge North America Inc Alaska Air Group Inc Union Planters Corp Tech Data Corp Energy East Corp Overseas Shipholding Group Flowserve Corp Urs Corp Corn Products Intl Inc Universal Compression Hldgs Northwestern Corp Solutia Inc Pinnacle West Capital

1510
4010 1010 2030 1010 2550 1010 2020 2530 2520 1510 3520 2550 1010 2550 4520 4510 4010 2510 4510 3520 4520 4520 1510 2550 2010 2550 2550 2550 4010 3510 2520 5510 2510 1510 2030 4010 4520 5510 1010 2010 2010 3020 1010 5510 1510 5510

Materials
Banks Energy Transportation Energy Retailing Energy Commercial Svcs & Supplies Hotels Restaurants & Leisure Consumer Durables & Apparel Materials Pharmaceuticals & Biotech Retailing Energy Retailing Technology Hardware & Equip Software & Services Banks Automobiles & Components Software & Services Pharmaceuticals & Biotech Technology Hardware & Equip Technology Hardware & Equip Materials Retailing Capital Goods Retailing Retailing Retailing Banks Health Care Equipment & Svcs Consumer Durables & Apparel Utilities Automobiles & Components Materials Transportation Banks Technology Hardware & Equip Utilities Energy Capital Goods Capital Goods Food Beverage & Tobacco Energy Utilities Materials Utilities

182

830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876

826 909 327 787 627 411 933 836 712 875 456 917 529 910 922 900 915 615 311 170 923 580 866 862 919 872 912 505 908 928 907 408 680 889 643 987 850 925 920 880 887 944 873 905 924 984 657

907 842 160 572 219 411 784 860 648 580 976 773 950 349 963 927 300 275 766 721 749 949 813 735 621 915 209 951 939 174 673 935 982 851 960 602 272 657 904 903 171 942

PBY DRRA DPH SCS IP ADM TWTC POL CKC TN13 SVU UNS CY DTG USG ARW OMX CEN AAPL HIG ALO TRI PAS YRK NC BFT AV OMG LPX SOV UHAL CSC MGM TSO CPWR AET NU PNM AM CQB CPC CMVT ASH GAP CUM HAL VSH

Pep Boys-Manny Moe & Jack Dura Automotive Sys -Cl B Delphi Corp Steelcase Inc Intl Paper Co Arch er-Dan iels-M idland Co Time Warner Telecom Inc Polyone Corp Collins & Aikman Corp Thomas & Betts Corp Supervalu Inc Unisource Energy Corp Cypress Semiconductor Corp Dollar Thrifty Automotive Gp Usg Corp Arrow Electronics Inc Officemax Inc Ceridian Corp Apple Computer Inc Hartford Fini Svcs Grp Inc Alpharma Inc -CIA Triad Hospitals Inc Pepsiamericas Inc York Intl Nacco Industries -Cl A Bally Total Fitness HIdg Cp Avayalnc Om Group Inc Louisiana-Pacific Corp Sovereign Bancorp Inc Amerco Computer Sciences Corp Metro Goldwyn Mayer Inc Tesoro Petroleum Corp Compuware Corp Aetna Inc Northeast Utilities Prim Resources Inc American Greetings A -CI Chiquita Brands Intl Central Parking Corp Comverse Technology Inc Ashland Inc Great Atlantic & Pac Tea Co Cummins Inc Halliburton Co Vishay Intirtechnology

2550 2510 2510 2020 1510 3020 5010 1510 2510 2010 3010 5510 4520 2030 2010 4520 2550 2020 4520 4030 3520 3510 3020 2010 2010 2530 4520 1510 1510 4010 2030 4510 2540 1010 4510 3510 5510 5510 2520 3020 2020 4520 1010 3010 2010 1010 4520

Retailing Automobiles & Components Automobiles & Components Commercial Svcs & Supplies Materials Food Beverage & Tobacco Telecommunication Services Materials Automobiles & Components Capital Goods Food & Drug Retailing Utilities Technology Hardware & Equip Transportation Capital Goods Technology Hardware & Equip Retailing Commercial Svcs & Supplies Technology Hardware & Equip Insurance Pharmaceuticals & Biotech Health Care Equipment & Svcs Food Beverage & Tobacco Capital Goods Capital Goods Hotels Restaurants & Leisure Technology Hardware & Equip Materials Materials Banks Transportation Software & Services Media Energy Software & Services Health Care Equipment & Svcs Utilities Utilities Consumer Durables & Apparel Food Beverage & Tobacco Commercial Svcs & Supplies Technology Hardware & Equip Energy Food & Drug Retailing Capital Goods Energy Technology Hardware & Equip

183

877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923

433 935 940 804 839 904 911 879 691 931 308 946 949 865 947 966 853 407 988 253 333 859 899 886 962 952 553 916 913 415 510 953 932 929 942 705 903 950 977 945 934 937 386 930 979 193 497

516 TE ANAT 996 AWGI 877 HUM 386 SPOT 975 BEV 757 TWR 554 NCR 414 USM 581 STEI 556 MLNM 763 SRP 566 TEN 908 AVT 474 CCI 398 AMT 637 DLM 165 TXT 231 GMST 970 NOC 561 TMPW 490 URI 865 IKN 555 PKS 279 DAL 583 ALV 310 EIX 811 FMC 328 CSX 630 ATIVIL 277 ODP 992 WLT 995 FL 969 PPE 492 BLI 417 TSN 614 ILA 991 AMD PVN 930 DDS 442 SKS 79 TLAB 172 BRCM 761 AKS 315 AMR 146 IPG 260 CC

Teco Energy Inc American National Insurance Alderwoods Group Inc Humana Inc Panamsat Corp Beverly Enterprises Tower Automotive Inc Ncr Corp Us Cellular Corp Stewart Enterprises A -Cl Millennium PharmactclsInc Sierra Pacific Resources Tenneco Automotive Inc Avnet Inc Crown Castle Intl Corp American Tower Corp Del Monte Foods Co Textron Inc Gemstar-TvGuide Intl Inc Northrop GrummanCorp Tmp Worldwide Inc United Rentals Inc Ikon Office Solutions Six Flags Inc Delta Air Lines Inc Autoliv Inc Edison International Fmc Corp CSX Corp Atmel Corp Office Depot Inc Walter IndustriesInc Foot Locker Inc Park Place Entmt Corp Big Lots Inc Tyson Foods Inc -Cl A Aquila Inc Advanced Micro Devices Providian FinancialCorp Dillards Inc -CIA Saks Inc Tellabs Inc BroadcomCorp -Cl A Ak Steel Holding Corp AMR Corp/De InterpublicGroup Of Cos Circuit City Stores Inc

5510
4030 3510 3510 2540 3510 2510 4520 5010 3510 3520 5510 2510 4520 5010 5010 3020 2010 2540 2010 2020 2550 4520 2530 2030 2510 5510 1510 2030 4520 2550 2010 2550 2530 2550 3020 5510 4520 4020 2550 2550 4520 4520 1510 2030 2540 2550

Utilities
Insurance Health Care Equipment & Svcs Health Care Equipment & Svcs Media Health Care Equipment & Svcs Automobiles & Components Technology Hardware & Equip Telecommunication Services Health Care Equipment & Svcs Pharmaceuticals & Biotech Utilities Automobiles & Components Technology Hardware & Equip Telecommunication Services Telecommunication Services Food Beverage & Tobacco Capital Goods Media Capital Goods Commercial Svcs & Supplies Retailing Technology Hardware & Equip Hotels Restaurants & Leisure Transportation Automobiles & Components Utilities Materials Transportation Technology Hardware & Equip Retailing Capital Goods Retailing Hotels Restaurants & Leisure Retailing Food Beverage & Tobacco Utilities Technology Hardware & Equip Diversified Financials Retailing Retailing Technology Hardware & Equip Technology Hardware & Equip Materials Transportation Media Retailing

184

924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970

969 938 901 88 967 948 601 603 358 321 968 959 536 939 941 963 821 970 195 783 961 640 957 964 199 94 951 955 926 936 943 974 971 490 975 392 973 346 985 522 989 782 98 164 983 956 972

968 627 235 42 946 994 342 137 730 233 353 130 699 464 399 622 997 105 258 420 488 508 989 191 41 771 337 331 543 359 217 497 114

89 999 49 60 244 379 152

X AW BRW F CCK 3FLMIQ LSI AGR. A CVS MWV UIS QTRN RTN PD CIVIS SRV CPN WBR KR GP CIEN AHC 01 IGL MU SUNW MON GR FD DCN TOY GT ADCT DYN EK LTR BBI WIMB MCH FON XRX CNP EP BA VC RRI CHTR

United States Steel Corp Allied Waste Inds Inc Broadwing Inc Ford Motor Co Crown Holdings Inc Fleming Companies Inc Lsi Logic Corp Agere Systems Inc Cvs Corp Meadwestvaco Corp Unisys Corp Quintiles Transnational Corp Raytheon Co Phelps Dodge Corp Cms Energy Corp Service Corp International Calpine Corp Wyndham Intl Inc Kroger Co Georgia-Pacific Corp Ciena Corp Amerada Hess Corp Owens-Illinois Inc IMC Global Inc Micron Technology Inc Sun Microsystems Inc Monsanto Co Goodrich Corp Federated Dept Stores Dana Corp Toys R Us Inc Goodyear Tire & Rubber Co ADC Telecommunications Inc Dynegylnc Eastman Kodak Co Loews Corp Blockbuster Inc Williams Cos Inc Millennium Chemicals Inc Sprint Fon Group Xerox Corp Centerpoint Energy Inc El Paso Corp Boeing Co Visteon Corp Reliant Resources Inc Charter Communications Inc

1510 2020 5010 2510 1510 3010 4520 4520 3010 1510 4510 3510 2010 1510 5510 3510 5510 2530 3010 1510 4520 1010 1510 1510 4520 4520 1510 2010 2550 2510 2550 2510 4520 5510 2520 4030 2550 5510 1510 5010 4520 5510 5510 2010 2510 5510 2540

Materials Commercial Svcs & Supplies Telecommunication Services Automobiles & Components Materials Food & Drug Retailing Technology Hardware & Equip Technology Hardware & Equip Food & Drug Retailing Materials Software & Services Health Care Equipment & Svcs Capital Goods Materials Utilities Health Care Equipment & Svcs Utilities Hotels Restaurants & Leisure Food & Drug Retailing Materials Technology Hardware & Equip Energy Materials Materials Technology Hardware & Equip Technology Hardware & Equip Materials Capital Goods Retailing Automobiles & Components Retailing Automobiles & Components Technology Hardware & Equip Utilities Consumer Durables & Apparel Insurance Retailing Utilities Materials Telecommunication Services Technology Hardware & Equip Utilities Utilities Capital Goods Automobiles & Components Utilities Media

185

971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000

79 987 THC 980 370 JCP 881 MIR 228 533 TXU 526 307 SANM 126 SLR 876 990 291 HLTH 273 220 S, 117 HI PRU 978 965 RJR ONE 593 UNM 976 142 ABS, 954 AWE 921 197 MCK 986 121 GLW` 991 FBF 249 65 HON 348 W13 992 324 VRSN 995 994 1000 KIND 996 234 GM 56 MOT 719 JPM 464 7 AOL 1000 12 LU 999 997 223 JDSU 26 SBC 993 22 T 998

Tenet HealthcareCorp Penney (J C) Co Mirant Corp Txu Corp Sanmina-SciCorp Solectron Corp WebMD Corp Sears Roebuck & Co Household InternationalInc Prudential FinancialInc RJ ReynoldsTobacco HIdgs Bank One Corp UnumprovidentCorp Albertsons Inc AT&T Wireless ServicesInc McKessonCorp Corning Inc FleetbostonFinancialCorp HoneywellInternationalInc Wachovia Corp Verisign Inc Kindred HealthcareInc General Motors Corp Motorola Inc JP Morgan Chase & Co AOL Time Warner Inc Lucent TechnologiesInc JDS UniphaseCorp SBC CommunicationsInc AT&T Corp

3510 2550 5510 5510 4520 4520 3510 2550 4020 4030 3020 4010 4030 3010 5010 3510 4520 4010 2010 4010 4510 3510 2510 4520 4020 2540 4520 4520 5010 5010

Health Care Equipment Svcs & Retailing Utilities Utilities TechnologyHardware& Equip TechnologyHardware& Equip HealthCare Equipment Svcs & Retailing DiversifiedFinancials Insurance Food Beverage& Tobacco Banks Insurance Food & Drug Retailing Telecommunication Services Health Care Equipment Svcs & TechnologyHardware& Equip Banks Capital Goods Banks Software& Services & HealthCare Equipment Svcs Automobiles& Components TechnologyHardware& Equip DiversifiedFinancials Media TechnologyHardware& Equip TechnologyHardware& Equip Services Telecommunication Services Telecommunication

186