Purpose: The main objective of this study is to examine the claim of Economic Value Added (EVA) proponents about its superiority as a financial performance measure compared to five traditional performance measures: net operating profit after tax (NOPAT), cash flow from operations (OCF), earnings per share (EPS), return on capital employed (ROCE) and return on equity (ROE) in Indian manufacturing sector and provide empirical evidences. To achieve this relative and incremental information content of various performance measures and their relationship with market value added (MVA) is tested and examined.
Design/Methodology: Principal components analysis (PCA) is one of the foremost multivariate methods utilized in business research for data reduction, latent variable modeling, multicollinearity resolution, etc. Our sample consists of 608 firms-year observations from the Indian manufacturing sector for a period of 2000-2007. Firstly, we employed Principal component Analysis (PCA) to determine the important variables that explains market value. Secondly, alongwith PCA multiple regression models (OLS) are used to examine the relative and incremental information content of EVA and traditional performance measures.
Findings: Our results about PCA reveal that variables like NOPAT, OCF, RONW, ROCE and EVA have maximum influence on the market value (MVA) of the sample companies, whereas EPS have negative loading so EPS is discarded for further analysis. Further, PCA loading matrix reveals that NOPAT, OCF, RONW and ROCE outscore EVA. Regression results about relative information content test reveals that NOAPT and OCF outperform EVA in explaining the market value of Indian companies. Incremental information content test show that EVA makes a marginal contribution to information content beyond NOPAT, OCF, ROCE and RONW. Overall our empirical results about Indian companies do not support the Stern- Stewart hypothesis that EVA is superior than traditional accounting based measures in association with market value of the firm.
Originality/value: The study concludes that alongwith financial variables other non- financial variables such employees, product quality etc. should be considered in order to capture the unexplained variation in the market value of the Indian companies.
Keywords: Return on capital employed, EVA, MVA, relative information content, incremental information test, PCA.
Paper Type: Research Paper
1. Introduction
Maximizing shareholders' wealth is well accepted objective among corporate mangers in recent times. The Corporate, that gave the lowest preference to shareholders curiosity are now bestowing the utmost preference to it. Shareholders' wealth is measured in terms of returns they receive on their investments (Sharma & Kumar, 2010). It can either be in forms of dividend or in the form of capital appreciation or both. Capital appreciation depends on the changes in the market value of the stocks. The market value of stocks depends upon number of factors ranging from company specific to market specific. Financial information is used by various stakeholders to assess firm's current performance and to forecast the future as well.
Any financial measures used in assessing firm's performance must be highly correlated with shareholders wealth and on the other hand should not be subjected to randomness inherent in it (Sharma and Kumar, 2010). The various empirical studies highlight that there is no single accounting measure which explains the variability in the shareholders wealth (Chen and Dodd, 1997; Rogerson, 1997). Traditional performance measures such as NOPAT, EPS, ROI, ROE etc. have been criticized due to their inability to incorporate full cost of capital thereby accounting income is not a consistent predictor of firm value and cannot be used for measuring corporate performance.
Value based management system has gained popularity in academic literature in last two decades. One such innovation in the field of internal and external performance measurement is Economic Value Added (EVA) [1] . Pioneered and advocated by US based business consultant Stern Stewart and company argue that economic value added can be used instead of earnings or cash from operations as measures of both internal and external performance. "Abandon earnings per share", "Earnings, earnings per share, and earnings growth are misleading measures of corporate performance" and "The best practical periodic performance measure is Economic Value Added (EVA)" (Stewart 1991).
The empirical studies such as Milunovich and Tsuei, 1996; O'Byrne, 1996; Uyemura et al., 1996; Biddle et al., 1997; Chen and Dodd, 1997, 2001; Bao and Bao, 1998; De Villiers and Auret, 1998; Turvey et al., 2000; Worthinton and West, 2001, 2004; Peixoto, 2002; Kyriazis and Anastasis, 2007; Maditinos et al., 2009 have been conducted in the last two decades, initially in the developed markets like USA and later in the rest of the international market , to answer if "it is really better to use value-based than traditional accounting performance measures to measure the financial performance of corporations, or which financial performance measure best explains corporations' change of value created and destroyed". However, the results reported by the studies are quite mixed and controversial.
The basic objective of this study is to examine the claim of Economic Value Added (EVA) proponents about its superiority as a financial performance measure compared to five traditional performance measures: net operating profit after tax (NOPAT), cash flow from operations (OCF), earnings per share (EPS), return on capital employed (ROCE) and return on equity (ROE) in Indian manufacturing sector and provide empirical evidences about developing market. For this purpose, relative and incremental information content of various performance measures and their relationship with market value added (MVA) is tested and examined. The structure of the study is as follows: the following section presents the brief account of review of literature, the next section describes the data and methodology used and the results of the statistical analysis are presented in the section before the conclusion of the study.
2. Literature Review
The empirical research for the value relevance of traditional accounting and modern value-based performance measures is broad but with controversial results (see Maditinos et al., 2009). As highlighted in the introduction section of the paper that there are various studies for both developed and developing markets to test the assertion that which corporate performance measures better explains the changes in shareholders' value. This section presents some of the prominent studies about the corporate performance measures.
2.1 Studies supporting the value based measures
Stewart (1991) provided evidence of the correlation between EVA and MVA. Using a sample US companies and examining both constant and changes in EVA and MVA, he found that there is a relationship between both the levels of EVA and MVA. Since the correlation between changes in EVA and MVA was high, Stewart suggested that adopting the goal of maximising EVA and EVA growth would in fact build a premium into the market value of the company. In a major study by Stern (1994) argues that the accounting measures such as earnings, earnings growth, dividends, dividend growth, ROE, or even cash flow are not key measures of corporate performance, but in fact EVA is one such measure that is closely linked with market value of company.
Milunovich and Tsuei (1996) investigated the correlation between frequently used financial measures (including EVA) and the MVA of companies in the US computer technology industry. The results clearly state that EVA demonstrated the best correlation and it would be fair to infer that a company that can consistently improve its EVA should be able to boost its MVA and therefore its shareholders' value. O'Byrne (1996) conclude that EVA explains more than twice as much of the variance in market/capital ratio as NOPAT when the EVA model has positive and negative EVA coefficient.. He also showed that EVA changes explain significantly more of the variation in market value changes.
Uyemura et al. (1996) studied the relationship between EVA alongwith traditional performance and MVA. They found that the correlation between MVA and those measures are: EVA 40 per cent, ROA 13 per cent, ROE 10 per cent, NI 8 per cent and EPS 6 per cent. The result concluded that EVA is better measures than ROA, ROE, NI and EPS in explaining the variation in market value of the companies. Lehn and Makhija (1997) examined the relationship between six performance measures and stock returns. The results revealed that EVA and MVA are effective measures of performance. Moreover, the correlation of EVA with stock returns (.59) was slightly higher than the correlation of MVA (.58), ROE (.46), ROA (.46), or ROS (.39). Thus, EVA and MVA appear to be somewhat better long-run performance measures than conventional accounting performance measures. Lee and Kim (2009) introduced Refined EVA (REVA) to the hospitality industry and compared it to EVA, market value added (MVA) and other traditional accounting measures (cash flow from operations (CFO), return on assets (ROA), and return on equity (ROE). The study provides interesting and meaningful findings that REVA and MVA can be considered good performance measures throughout the three hospitality sectors (i.e., hotel, restaurant and casino). According to the findings, REVA and MVA significantly explain the market adjusted return by presenting positive coefficients.
2.2 Studies supporting superiority of Traditional corporate performance measures
Biddle et al. (1997) tested the assertions that EVA is more highly associated with stock returns and firm's value than accrual earnings, and evaluated which components of EVA, if any, contributed to these associations. The results indicated that earnings (R2 =12.8%) were significantly associated with market adjusted annual returns than either Residual Income (R2 = 7.3%) or EVA (R2 = 6.5%) and that all three of these measures dominate cash from operations (R2 =2.8%). The empirical results regarding relative information content, rather suggest that earnings generally outperform EVA. Similar results were revealed by Kramer and Pushner (1997) by analyzing the strength of the relationship between EVA and MVA, using the Stern Stewart 1000 companies for the period between 1982 and 1992.They found that although MVA and NOPAT were positive on average, the average EVA over the period was negative. No clear evidence is found to support the contention that EVA is the best internal measure of corporate success in adding value to shareholders' investment.
Chen and Dodd (2001) empirically examined the value-relevance of three profitability measures- Operating Income (OI), Residual Income (RI), and Economic Value Added (EVA) and concluded that the market may place higher reliance on audited accounting earnings than the unaudited EVA metric. Their findings failed to support the assertion that EVA is the best measure for valuation purposes. Moreover, Worthington and West (2001) in their study about Australian companies clearly suggested the superiority of EVA compared to earnings and other accounting performance measures in explaining stock returns.
Ismail (2006) in a study about UK companies tested the relative and incremental information content of EVA and other performance measures using panel data regression. The results of the study fail to support the Stern- Stewart hypothesis as net operating income after taxes and net income outperform EVA and residual income. The paper concludes that apart from financial variables other factors like employee, customer satisfaction and R&D initiatives must be considered to capture the changes in the stock return.
Similarly, Kim (2006) provides empirical evidence on the relative and incremental information content of EVA and traditional performance measures, earnings, and cash flow of hospitality industry of U.S. The information content of EVA and other explanatory variables indicate that earnings are more useful than cash flow in explaining the market value of hospitality firms. Kyriazis and Anastassis (2007) investigated the relative explanatory power of the Economic Value Added (EVA) model with respect to stock returns and firms' market value. They conclude that net and operating income (NOPAT and OP) appear to be more value relevant than EVA in explaining the market value of firms listed at Athens Stock Exchange (ASE).
Ismail (2008) provide evidence regarding Economic Value Added (EVA) and company performance in Malaysia. The study sought to explain the ability of EVA, compared to traditional tools, in measuring performance under various economic conditions; pre-economic crisis, during economic crisis and post-economic crisis period. The result of the study found that traditional tools particularly EPS is able to correlate and had a relationship with stock returns. Further the study revealed that EVA is also able to correlate with stock returns and is superior in explaining the variations in the stock returns as compared to the traditional tools under varying economic conditions.
Maditinos et al. (2009) examined the explanatory power of two value-based performance measurement models, EVA and SVA, compared with three traditional accounting performance measures: earnings per share (EPS), return on investment (ROI), and return on equity (ROE), in explaining stock market returns in the Athens Stock Exchange (ASE). The results of relative information content tests reveal that stock market returns are more closely associated with EPS than with EVA or other performance measures. However, incremental information content tests suggest that the pairwise combination of EVA with EPS increases significantly the explanatory power in explaining stock market returns.
Many other studies reported the weak correlation of residual income based metrics with shareholders value as measured by stock returns. Like, Peterson and Peterson (1996) provided evidence that EVA and other valued based type measures do not provide much more information than stock prices. Stark and Thomas (1998) examined the UK market and concluded that the relationship between residual income (RI) and market value is by no means perfect. Goetzmann and Garstka (1999) in their study that long-term survival of companies may be related to accounting earnings, and more, simple EPS does as well or better than EVA at explaining differences across companies and at predicting future performance.
3. Data and Methodology
3.1 Sample section
Our sample period spans 08 years, from 2000 to 2007.Sample size consist of 76 companies from India and listed on Bombay Stock Exchange (BSE). The selected companies are from manufacturing sector (as defined by Business Beacon -CMIE database) and taken from BT-500 companies (India's most valuable companies, Business Today, 2006). The rationale behind selecting BT-500 as sample base is that these companies are ranked on the basis of market capitalization in the Indian stock market and hence, can be considered as India's most valuable companies. Initially 104 companies (Kaur and Narang, 2009) with full EVA data from 2000-2007 were identified from the BT- 500 companies. Then, we excluded 28 firms which are not covered in manufacturing companies as per CMIE database. So, a final sample of 76 companies with full data from 2000-2007 are included forming 608 firm- years observations.
3.2 Variables and data source
The main objective of the study is to examine the relative and information content of EVA and traditional performance measures. To achieve this, Market Value Added (MVA) is defined as the value added in excess of economic capital employed and used as dependent variable. In order to calculate the economic capital we made certain adjustments as proposed by Stern- Stewart & company. These include non- recurring gains/losses, capitalized R&D expenses etc. Along with MVA, EVA, NOPAT, OCF, RONW, ROCE and EPS are used as independent variables. The rationale for selection of these variables is that the same are used by majority of the researchers all over the world. Data used in the present study is taken from Capitaline database and data related to EVA is taken from the study conducted by Kaur and Narang, 2009.
3.3 Hypothesis of the study
We examined the usefulness of various performance measures in terms of their ability to explain the variation in the shareholders value. Based on the objective of the study, following hypotheses are formulated:
Hypothesis 1: The relative information content of EVA is superior to traditional performance measures (NOPAT, RONW, ROCE, EPS and OCF) in explaining market value of Indian companies.
Hypothesis 2: EVA adds more information content beyond the NOPAT, RONW, ROCE, EPS and OCF in explaining market value of firms.
4. The Model Specification
In order to examine the relative information content and incremental information content of various performance measures, we used two level of analysis: Factor analysis (Principal component analysis) and pooled regression (OLS). Principal component (PCA) involves the calculation of the Eigen value decomposition of a data covariance matrix or singular value decomposition of a data matrix, usually after mean centering the data for each attribute. The results of a PCA are usually discussed in terms of component scores and loadings (Shaw, 2003).
Although, basic application of PCA is (1) to get a small set of variables from a large set of variables (most of which are correlated to each other) and (2) to create indexes with variables that measure similar things (conceptually) but the basic purpose of application of PCA in the present paper is to find out the number of factor we can include for OLS regression analysis. The variables with negative loading are discarded from the analysis and excluded from the model formation. In second step, we used univariate and multivariate regression (OLS) to test the relative information content and to find out the relationship of various performance measures with Market value added (MVA). More specifically following regression models are used in the present study.
Univariate Regression Models: To test the Relative Information content of each variables
MVAit =β0 + β1EVAit + εit …………………….. (1)
MVAit =β0 + β1NOPATit + εit ……………………. (2)
MVAit =β0 + β1ROCEit + εit ……………………..(3)
MVAit =β0 + β1RONWit + εit ……………………. (4)
MVAit =β0 + β1OCFit + εit ……………………….(5)
Multiple Regression Models: To Test the Incremental information content of EVA and traditional financial performance measures
MVAit =β0 + β1NOPATit+ β2ROCEit+ β3 RONWit + β4 OCFit + εit
……………… (6)
MVAit =β0 + β1EVAit+ β2 NOPATit + β3 ROCEit + β4 RONWit + β5 OCFit + εit
……………… (7)
Where: MVAit ,amount of market value added for the firm i in period t as above ; EVAit, amount of economic value added of firm i in period t; NOPATit, net operating profits after taxes for firm i in period t; ROCEit, ratio of earning before taxes to capital employed for firm i in period t; RONWit, ratio of net income after tax to networth for firm i in period t; OCFit cash flow from operations for firm i in period t; εit ,stochastic error term for firm i at time t ; and i= 1,…….76 and t= 1,……8.
5. Results and Analysis
5.1 Sample adequacy and validity test
Undertaking a good factor analysis relies on the sample adequacy and validity. A very important validity test criterion for factor analysis is the Kaiser-Meyer Olkin - KMO (after Kaiser, 1974). This criterion is used to measure the sample adequacy by comparing the magnitude of the observed partial correlation coefficients to those of the magnitude of the partial correlation coefficients. Kaiser (1974) suggested that each attribute must have a high KMO coefficient score as this improves the value of each individual attribute. According to Kaiser (1974); Bryman and Cramer (1999) a KMO value under 0.50 is perceived as being inadequate and therefore unacceptable. The value of KMO statistics of our sample as shown in Table I, KMO = .579, which is above the specification of 0.50. Further, Bartlet test of Sphericity for adequacy (strength of the relationship among variables) of correlation matrix is significant as the value is 0.000 which is less than 0.05 revealing that correlation matrix is not an identity matrix. So, the results of KMO and Bartlett's test reveal that our sample is fit for factor analysis.
Table: I Results of KMO and Bartlett's test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy.
.579
Bartlett's Test of Sphericity
Approx. Chi-Square
507.149
df
21
Sig.
.000
5.2 Descriptive Statistics and Correlation Matrix
Panel A of Table II provides the descriptive statistics of variables used in the study. It is clear from the table that RONW (12.17809) and ROCE (17.19) have the lowest standard deviation standard deviation among the independent variables, followed by EPS (53.32057) and EVA (241.82620). MVA, NOPAT and OCF reveal the highest standard deviation. Mean statistics show that MVA, NOPAT and OCF are negative. Negative MVA implies that most of the sample companies are destroying value as against the value creation for shareholders. Highest mean value of EVA (39.3640) implies that most of the sample companies are adding value to shareholders' wealth.
Table II: Descriptive Statistics
Panel A: Descriptive Statistics
MVA
EVA
NOPAT
EPS
OCF
ROCE
RONW
Mean
3.9407E3
39.3640
3.7702E2
35.1612
4.6391E2
27.5982
23.8222
S.D
9744.89969
241.82620
906.52728
53.32057
1268.97934
17.19106
12.17809
Minimum
-349.95
-456.12
28.73
5.68
23.31
5.76
4.77
Maximum
64264.81
1211.99
7288.75
359.49
10467.66
114.44
82.50
Panel B: Correlation Matrix
MVA
EVA
NOPAT
EPS
OCF
ROCE
RONW
EVA
.392**
1
NOPAT
.474**
.132
1
EPS
-.024
.034
.051
1
OCF
.418**
.048
.990**
.056
1
ROCE
.230*
.499**
.037
.036
-.003
1
RONW
.280*
.528**
.083
-.031
.042
.916**
1
MVA
1
Note: - **, *. Correlation is significant at the 0.01 and 0.05 level.
5.3 Principal Component Analysis
The aim of principal component analysis is the construction of a set of variables "Xi"s (i= 1,2…..k), of a new set of variables (Pi) called principal components, which are linear combinations of the "X"s. These combinations are chosen so that the principal components satisfy two conditions (Gupta and Gangadhar, 2004):
the principal components are orthogonal to each other, and
the first principal component accounts for the highest proportion of total variation in the set of all "X"s, the second principal component accounts for the second highest proportion and so on.
By applying six independent variables (NOPAT, ROCE, RONW, EPS, OCF and EVA) to PCA we get percentage variance with respect to Principal Components (PC) as mentioned in Table-III. Selecting only top 3 PCs with Eigen values more than 1, which for giving nearly 89% of the variance explained, there loadings are considered for finding the influence of the variable.
Table III. Results of PCA showing PCs with Eigen values and corresponding variance
Principal
Components
Eigen
Value
% Variance
Cumulative
variance
1
2.366
39.43
39.43
2
1.953
32.55
71.98
3
1.001
16.63
88.61
4
0.595
9.95
98.56
5
0.0793
1.34
99.90
6
0.006
0.10
100.0
The loadings obtained for different Principal components are given in Figure 1 to 3 (see Annexure-I). The values of individual loadings of different PCs are also shown in Table IV. The sum total values of different variables shows that EPS has negative influence as we are selecting only 3 PCs with Eigen value of greater than 1 for our analysis. Table V summarizers the loading of different variables as per priority for further analysis and we found the variables like NOPAT, OCF, RONW, ROCE and EVA has maximum influence, whereas EPS have negative loading. Thus, EPS is discarded for further analysis.
Table IV. Loadings of different variables for top 3 PCs
Variables
PC1
PC2
PC3
Total
EVA
0.467
-0.1185
-0.0435
0.305
NOPAT
0.2573
0.6546
0.0558
0.9677
EPS
0.03173
0.06862
-0.9941
-0.8937
OCF
0.221
0.6694
0.0528
0.9432
ROCE
0.5697
-0.2423
-0.0165
0.3109
RONW
0.5842
-0.2142
0.0603
0.4303
Table V. Loadings of different variables as per priority
Variables
PC1
PC2
PC3
Total
NOPAT
0.2573
0.6546
0.0558
0.9677
OCF
0.221
0.6694
0.0528
0.9432
RONW
0.5842
-0.2142
0.0603
0.4303
ROCE
0.5697
-0.2423
-0.0165
0.3109
EVA
0.467
-0.1185
-0.0435
0.305
EPS
0.03173
0.06862
-0.9941
-0.8937
5.4 Results of Regression Analysis
In Table VI, we report the results of relative information content test. The assessment is made by analyzing univariate regression results of independent variables i.e. EVA, NOPAT, ROCE, RONW and OCF. Earning per share (EPS) as discussed above is not included for regression analysis. Results are estimated on the basis of regression equations (1) to (5). In order to test the incremental information content of performance measure, we compared the change in the value of adjusted R2 from regression model 6 to 7. Before examining the regression results we also test the first order serial correlation and multicollinearity in our data. Serial correlation was analyzed by examining the Durbin-Watson (D-W) statistics. Since our maximum value of D-W statistics was found to be 1.935 which is in range (1.5-2.00) indicating no possible autocorrelation. In order to check the independence of the predictor variables (independent), Variance Inflation factors (VIF) was analyzed. We found 3.81 as the highest value of VIF about EPS and thereby concluding no colineraity in the data. Following are the some of the important observations from the OLS regression models:-
From the relative information content test, we found that NOPAT and OCF outperform EVA as coefficient of determination ( R2) about EVA is 15.4 percent as compared to 22.5 percent and 17.5 percent for NOPAT and OCF respectively. This means that traditional performance measures explain better in the variation in MVA of the sample companies. Regression results further reveals that all the individual coefficients (β) are significant as t- statistics are significant at 0.05 level of significance. So, our hypothesis (H1) that relative information content of EVA is superior to traditional measures is rejected and thereby concluding that traditional corporate performance measures are better indicator of changes in market value of Indian Companies.
Our results about relative information content test are consistent with majority of international studies such as (Chen and Dodd, 1997; Biddle et al., 1998; Ray, 2001; Worthington & West, 2001; Peixoto, 2002; deWet, 2005; Ismail, 2006; Kim, 2006; Kyriazis and Anastassis, 2007; Vijayakumar and Selvi, 2007; Visaltanachoti et al., 2008; Maditinos et al., 2009) but different from many studies (Irla, 2007; Sunitha, 2008, Taufik et al., 2008).
Our results about Incremental information content test reveal that EVA add marginal (.009) in explaining the market value of the firm as compared to without considering EVA. OCF and ROCE exhibit negative relationship with MVA of the sample companies. Overall model results (as revealed by F statistics) are significant with statistically significant p value (.000). Thus our second hypothesis (H2) that EVA adds more information content to that provided by NOPAT, RONW, ROCE, EPS and OCF in explaining market value of firms is rejected.
Further regression results reveals that only 35(adj. R2) percent of variation in MVA is explained by the regression model leaving majority unexplained. It means that apart from these financial variables one should consider non financial variables like customer satisfaction, product quality to capture the exact variation in MVA of the sample companies
Overall, our results failed to support the Stern- Stewart hypothesis that EVA is a better performance measures. Traditional performance measures such as NOPAT, OCF are better than EVA. On the line of Chen and Dodd (2001) and Ismail (2008) we also feel that one should consider non -financial variables to capture the changes in market value of the company.
Table VI: Regression Results of Relative Information Content
Rank order of R2
(1)
(2)
(3)
(4)
(5)
All firms
NOPAT
>
OCF
>
EVA
>
RONW
>
ROCE
R2 (percent)
22.5
17.5
15.4
7.8
5.30
Adj. R2 (percent)
21.5
16.4
14.3
6.6
4.00
Coefficients
.573(4.637)*
0.519(3.961)*
0.408(3.670)*
0.269(2.508)**
0.22(2.037)**
P-value
0.000
0.000
0.000
0.014
0.045
F
21.497*
15.689*
13.470*
6.288*
4.148
Note: - *, ** coefficients are statistically significant at 0.01 and 0.05 respectively level of significance.
Table VII: Test Results of Incremental Information Content
Independent variables
Model 1
Model 2
RONW
.209(.917)
.170(.735)
NOPAT
3.301(3.929)*
2.760(2.770)**
ROCE
-.072(-.321)
-.082(-.367)
OCF
-2.851(-3.309)*
-2.307(-2.270)**
EVA
-
.137(2.449)**
R2
.379
.388
Adjusted R2
.344
.345
F-value
10.850*
8.885*
∆ R2
-
.009
Durbin-Watson
1.935
1.875
Notes: MVA-measure market value added; EVA- economic value added; RONW- return on net worth; NOPAT-net operating profit after taxes; ROCE- return on capital employed; OCF- operating cash flows; t-statistics are provided in parenthesis; *, **statistically significant at 1 and 5 percent level respectively.
6. Concluding Remarks
The basic objective of this study is to examine the claim of EVA proponents about its superiority over traditional performance measures in explaining the variation in the shareholders' value in Indian market and provide empirical evidence. Using data set of 76 Indian manufacturing companies for time span from 2000-2007, we test the relative and incremental information content test of EVA and conventional performance measures. We used two step methodologies, i.e. factor analysis and OLS regression. The basic rationale of using factor analysis was to determine the number of factors that we can use for further analysis. Principal component analysis (PCA), popular form of factor analysis was used and result about PCA reveal that since Earning per share (EPS) has negative value for factor loading so can be discarded for regression analysis. The regression results about relative information content test reveal that NOPAT and OCF outperform EVA and thereby rejecting the hypothesis that EVA has better explanatory power than conventional performance measures. Our incremental information content test further reveals that EVA makes only marginal contribution and has no significant impact on the changes in the market value of the sample companies. The relatively low explanatory power of all performance measures under examination is largely consistent with the results of many international studies (see Chen and Dodd, 1997, 2001; Biddle et al., 1997; Maditinos, 2009).The result of the present study support the claims of many researchers that more financial and non- financial determinants should be employed to assess the value of the firm. We have identified following areas for future research to examine the relevance of various corporate performance measures in explaining the market value of Indian companies:
To test the data using alternative dependent variables (such as stock returns) with same or large data set of similar or more time span. Many international studies have used stock returns instead of MVA to examine the value relevance of various performance measures.
To test the incremental information content of components of EVA ( such as Adjusted NOPAT, Cost of Capital, Economic capital and Accounting adjustments) with similar or large data set.
To examine the value relevance of financial and banking institutions and make comparison with results of other sectors.
Selected References
Bao, B-H. & Bao, D-H. (1998). Usefulness of value added and abnormal economic earnings: An empirical examination', Journal of Business Finance and Accounting, 25(1-2), 251-65.
Biddle, G. C., Bowen, M.R. and Wallace, J.S. (1998). "Economic Value Added: Some Empirical EVAdence". Managerial Finance, 24(11): 60-71.
Bryman, A. and Cramer, D. (1999), Quantitative Data Analysis with SPSS Release 8 for Windows. A Guide for Social Scientists, Routledge, London.
Business Today (2006). BT-500 India's Most Valuable Companies. Special Issue, November 19, 118-138.
Business Today. (2000). Just How Much Wealth Does Your Company Create?. BT-Stern Stewart Research Project, February 22 to March 6, 81-140.
Chen, S. and Dodd, J. L. (1997). Economic Value Added: an empirical examination of a new corporate performance measure. Journal of Managerial Issues, 9(3): 318-333.
Chen, S. and Dodd, J.L. (2001). "Operating income, residual income and EVA: which metric is more relevant?". Journal of Managerial Issues, 13:65-86.
De Villiers, J. and Auret, C.J. (1998). A comparison of EPS and EVA as explanatory variables for share price. Studies in Economics and Econometrics, 22(2): 47-63.
DeWet, J.H. (2005). EVA versus traditional accounting measures of performance as drivers of shareholder value - A comparative analysis. Meditari Accountancy Research, 13(2): 1-16.
Goetzmann, W. N and S. J. Garstka (199). The Development of Corporate Performance Measures: Benchmarks Before EVA, Yale ICF Working Paper, 99-06, July 12, New Haven; Yale School of Management.
Gupta, D and Gangadhar, B.R. (2004). A statistical model for developing body size charts for garments. International Journal of Clothing Science and Technology, 16(5):458-469
Irla, L.R. (2007). Corporate performance measures in India: An empirical analysis. Working papers no WP-2007/01/A. [Online]. Available: papers.ssrn.com/sol3/Delivery.../SSRN_ID967664_code709346.pdf?
Ismail, A. (2006). Is economic value added more associated with stock return than accounting earnings? The UK evidence. International Journal of Managerial Finance, 2(4): 343-353.
Ismail, I. (2008). Performance of public-listed companies in Malaysia: using EVA. Paper presented at the 6th One- day Symposium on Accountability, Governance and Performance, 15 February, Brisbane, Australia, available at: www.griffith.edu.au/school/gbs/afe/symposium/2008/Ismail.pdf.
Kaiser, H.F. (1974), An index of factorial simplicity, Pschometrika, 39: 31-6.
Kaur, M. and Narang, S. (2009). Shareholder value creation in India's most valuable companies: an empirical study. The IUP Journal of Management Research, VIII (8):16-42.
Kim, W. (2006). "EVA and traditional accounting measures: which metric is a better predictor or market value of hospitality companies?", Journal of Hospitality & Tourism Research, 30(1): 34-49.
Kramer, J. K. and Pushner, G. (1997). An Empirical Analysis of Economic Value Added as a Proxy for Market Value Added. Financial Practice and Education, 7(1): 41-49.
Kyriazis, D. and Anastassis, C. (2007). The validity of the economic value added approach: an empirical application. European Financial Management, 13(1):71-100
Lee, S. and Kim, W. G. (2009). EVA, refined EVA, MVA, or traditional performance measures for the hospitality industry. International Journal of Hospitality Management, 28(3), 301-484.
Lehn, K. & Makhija, A. K. (1997). EVA, accounting profits, and CEO turnover: an empirical examination 1985-1994. Journal of Applied Corporate Finance, 10(2), 90-96.
Maditinos, D.I, Sevic, Z. Theriou, N. G. (2009). Modelling traditional accounting and modern value-based performance measures to explain stock market returns in the Athens Stock Exchange (ASE). Journal of Modelling in Management, 4(3): 182-201
Milunovich, S. & Tsuei, A. (1996). EVA in the computer industry. Journal of Applied Corporate Finance, 9(1):104-115.
O'Byrne, S.F. (1996). EVA and Market Value. Journal of Applied Corporate Finance, 9(1): 116-125.
Peixoto, S. M., (2002). Economic Value Added - application to Portuguese Public Companies. [Online]. Available at SSRN: http://ssrn.com/abstract=302687 or doi:10.2139/ssrn.302687. (Retrieved on July, 10, 2009)
Peterson, P. P. & Peterson, D.R. (1996). Company Performance and Measures of Value Added. The Research Foundation of the Institute of Chartered Financial Analysts, Charlottesville, VA.
Rogerson, W.P. (1997). Intertemporal cost allocation and managerial investment incentives: A theory explaining the use of economic value added as a performance measure", Journal of Political Economy, 105(4): 770-95.
Sharma, A. K. and Kumar, S. (2010). Effectiveness of EVA and conventional performance measures: Evidences from India. IIMS- Management Science, 1(1), 60-78.
Stark, A.W. and Thomas, H.M. (1998) .On the empirical relationship between Market Value and Residual Income in the U.K., Management Accounting Research, 9: 445-460.
Stern, J.M. Stewart, G.B. III and Chew, D.H. (1994). The EVA financial management systems. Journal of Applied Corporate Finance, 7(2): 32-46.
Stewart III, G.B., 1991. The Quest for Value. New York, Harper Business.
Sunitha, K. (2008). Innovative versus traditional financial metrics: an appraisal. Globsyn Management Journal, 2(1):19-26.
Taufik, H., Isnurhadi¸H and Widiyanti, M. (2008). The influence of traditional accounting and economic value added approaches on stock returns of banks listed on Jakarta Stock Exchange (JSX). Proceedings of The MFA Conference 2008, Available at: digilib.unsri.ac.id/download/SESSIONIIB.pdf (Retrieved on July, 15, 2009)
Tuvey, C.L., Van Duren, E. & Sparling, D. (2000). The relationship between economic value added and the stock market performance of agribusiness firms. Agribusiness, 16(4):399-416.
Uyemura, D.G., Kantor, C.C. and Petit, J.M. (1996). EVA for banks: value creation, risk management and profitability measurement", Journal of Applied Corporate Finance, 9(2), 94-111.
Vijayakumar, A. and Manor Selvi, A. (2007). EVA and MVA of Indian automobile industry - an empirical study of relationship. Abhigyan, 25(3): 54-65.
Visaltanachoti, N., Luo, R. and Yi, Y (2008). Economic value added and sector returns. Asian Academy of Management Journal of Accounting and Finance, 4(2): 21-41.
Worthington, A. & West, T. (2004).Australian Evidence Concerning the Information Content of Economic Value-Added. Australian Journal of Management, 29(2), 201-224.
Worthington, A. and West, T. (2001). The Usefulness of Economic Value-Added (EVA) and its Components in the Australian Context. Accounting, Accountability and Performance, 7(1): 73-90.
Annexure- I Factor Loading of Principal components (PCs)
Figure. 1. Loadings of PC-1
Figure 2. Loadings of PC-2
Figure 3. Loadings of PC-3
