# chp_10.1007_978-3-642-40063-6_68.docx

Study on the Perance uation of Manufacturing Enterprises Based on EVA Zhi-gang Li, Xi Zhao, Xu Gong, and Ke-jia Liu Abstract Focusing on the manufacturing enterprises in growth enterprises market board, this paper adopt the EVA index into the perance uation of those companies, using empirical analysis to confirming the feasibility. Next, a comparison and relationship was made between the EVA and the MVA. A conclusion can be drawn that EVA has a strong ability of explanation of MVA and the feasibility of using EVA to uate the perance of those enterprises. For the EVA, the capacity of explanation of the MVA is superior to the traditional perfor- mance uation index. In addition, combining EVA with traditional perance measures, it can a more comprehensive and reasonable perance uation . Keywords EVA MVA Perance uation 1 Introduction Perance uation has been a hot topic in theory and practice discussion, from the DuPont Analysis on investment to a series of indicators for perance uation. There have been many financial indicators to uate the perance of companies nowadays, especially for the manufacturing enterprises. Stern Chen and Dodd 1997. In addition, the EVA is better than traditional financial index in the ability of reflecting the MVA Lehn and Makhija 1996a, b. 2 Case Study 2.1 Sample and Index Samples are chosen from companies listed on the board of Shenzhen Stock Exchange from 2009 to 2010. Given the short-sighted behavior after listed com- panies listed and instability, in order to eliminate the volatility and the impact of financial instability, this paper selects listed with more than 1 year and can obtain nearly complete financial data of the enterprise as the research sample. As a result, 2009 samples of listed companies are 36. And 58 companies in 2010 data samples, the following empirical analysis in the sample size are 94. Most of the entire sample is the manufacturing enterprises. The data is elected respectively from WIND consulting database, and the website of Peoples Bank of China. Use SPSS18.0 version as analysis software for data processing. 2.2 Variable Calculation In order to get the EVA of every company, this paper chooses some indicators calculate it, NOPAT net operating profit after tax, WACC weighted average cost of capital, TC total capital. In addition, we also need MVA added value of the market of all enterprises. And, the r is for the correlation coefficient. Calculation process is as follows 2.2.1 NOPAT After-tax profits for the appropriate accounting adjustment, the adjustment ula is as follows Study on the Perance uation of Manufacturing Enterprises Based on EVA 687 qX NOPAT D after-tax net profit C interest payments exchange gain C .non business expenses non operating income/ .1 income tax rate/ C deferred income tax liabilities increase deferred income tax assets increase 2.2.2 TC Total capital total liabilities and owners equity, not in the financial statements and should be properly adjusted, in this paper, combined with the database data and ination provided, the total capital properly accounting adjustments Total capital D owner0s equity C debt projects under construction notes payable-accounts payable in advance 2.2.3 WACC WACC D Rs Ws C Rd Wd .1 T / Among them, Rs is equity cost of capital rate, Ws is the proportion of equity capital, Rd is the debt capital cost rate, Wd is the proportion of debt capital, T is income tax. 2.2.4 MVA 1 MVA D XEVAi i Among them, r is the discount rate. 2.2.5 Correlation Coefficient i D1 .1 C r / Correlation analysis is mainly studying the relationships between variables. Correla- tion coefficient is can measure variables related degree between inds. This paper uses the Pearson correlation coefficient measuring the correlation between variables to compare the ability of exploration of the traditional perance uation index with that of MVA and EVA value. Pearson correlation coefficient usually use letters, its computation ula is X .xi x/ .yi y/ r D .xi x/2.yi y/2 688 Z. Li et al. Among them, xi is the various perance indicators variables, yi is MVA value, r is the correlation coefficient and correlation degree between two variables. The scope is 1, 1, the bigger the jrj, the stronger of the correlation between the two variables is. 2.3 Hypothesis EVA is from enterprise itself fundamentals to uate company perance indicators, can be used for the real measure of the status of the enterprise to create wealth for shareholders. MVA, the companys market value, has the ability to reflect the expectation for the capital market to the added value of the listed company access to future EVA. If the market is efficient, the market value and market price should be consistent; the present value of the company stock price can reflect this value. This is to say, enterprises intrinsic value and market value match degree is higher, the correlation of MVA and EVA is also stronger. 1. The companies EVA and MVA highly relevant, of MVA and EVA have strong explanation ability. Have been proposed in the theoretical research part of this paper, the tradi- tional perance uation s did not consider the equity capital cost, to some extent, distort the true operating perance of enterprise. While, EVA is based on economic profit, taking all the cost of capital, including equity capital into account and through accounting adjustment to eliminate the ination distortion existing in the current accounting standards. EVA embodies a period for shareholders to create or damage to the value of the real as defined by the profits for shareholders of the enterprise. 2. For the company, EVA for MVA explanation ability than traditional perance uation index. 3 Result Use Pearson correlation coefficient analysis to analyze the relationship between the various perance indicators MVA, EVA, NI, ROE, EPS and CFOPS of the sample company Table 1. Correlation coefficient of MVA and EVA is higher than that of the other perance uation index correlation coefficient. To some extent, it justifies the hypothesis 1 and hypothesis 2. According to hypothesis 1, explained by the added value of MVA of market variables, EVA set up linear regression model as explained variables. To regression analyze the 58 listed companies from 2009 to 2010 a total of 94 samples, a linear regression model is established as follows MVA D C EVA C Study on the Perance uation of Manufacturing Enterprises Based on EVA 689 Table 1 The correlation coefficient of variables Variable MVA EVA NI ROE EPS CFOPS MVA 1 EVA 0.873** 1 NI 0.934** 0.956** 1 ROE 0.633** 0.721** 0.710** 1 EPS 0.239* 0.276** 0.395** 0.366** 1 CFOPS 0.212* 0.177* 0.240* 0.238* 0.581** 1 Note ** under 0.01 level correlation significantly two-tailed test, * showed significant correlation under 0.05 level two-tailed test Table 2 EVA and MVA regression results Model summary DW ANOVA Coefficients Independent variables R2 Adjusted R2 DW F Sig. Constant Coefficient t-value Sig. EVA 0.745 0.742 1.461 268.821 0.000 1.733E C 09 47.973 16.396 0.000 For the variable t-test, the P values significantly under less than the significance level of 0.001, shows that the variables pass through the t-test. The equation P values significantly under less than the significance level of 0.001, shows that the regression equation pass through F -test. From the view of goodness of fit, EVA explanation for MVA degree is higher, the adjusted R2 reached 74.2 Table 2, shows that EVA variables in equation of dependent variable MVA strong degree reached 74.2 . To eliminate autocorrelation, through the Durbin Watson d inspection test whether variables exist autocorrelation, DW value of 1.461, significantly closer to 2, and indicates that there is no serial correlation. The explanation of EVA for MVA has strong ability, thus hypothesis 1 confirmed. At the same time, we also observed the adjusted R2 is 74.2 , which shows that EVA is not enough to fully explain the MVA and EVA indicators can serve as the important inds for uation of enterprise business perance, but only to EVA as the enterprise operating perance uation index is still incomplete. Next, to confirm the hypothesis 2, there are two kinds of s. The first is computed for each variable incremental ination content, namely for make each perance index variables and MVA for a linear return, then put EVA variables into each model, making the er model to a bivariate regression model, and make a comparison the interpretational ability of the regression equation of the before one and after one, namely fixed R2 values. The second is establishing multiple regressions, respectively with NI net income, ROE rate of return on common stockholders equity, EPS earnings per share and CFOPS operating cash flow per share as independent variables, with the added value of market MVA as dependent variable. Then add variables EVA to the multivariate regression model, then compared the interpretational ability of the regression equation of before one and after one. 690 Z. Li et al. Adjusted R2 R2 Table 3 Traditional indicators and MVA regression results Model summary Independent variable ANOVA Coefficients F Sig. t-value Sig. NI 0.673 0.672 333655 0.000 3.914E C 08 48.021 25173 0.000 ROE 0.401 0.395 61671 0.000 2.500E C 09 7.621E C 10 7853 0.000 EPS 0.057 0.047 5590 0.020 2.440E C 09 1.91E C 09 2364 0.020 CFOPS 0.037 0.027 3539 0.063 3.306E C 09 1.15E C 09 1881 0.063 Table 4 Traditional indicators after the introduction of EVA and MVA regression results Model summary ANOVA Coefficients Independent Adjusted variable R2 R2 F Sig. Constant term Coefficients T-value Sig. VIF NI EVA 0.883 0.881 344.804 0.000 1.339E C 07 64.848 10394 19.046 2823 0.000 11506 0.006 ROE EVA 0.745 0.740 133.140 0.000 1.539E C 09 2.859E C 09 0311 47.021 11086 0.756 2081 0.000 EPS 0.745 0.739 132.950 0.000 1.728E C 09 9.265E C 06 0021 0.983 1082 EVA 47.955 15667 0.000 CFOPS EVA 0.747 0.741 134.097 0.000 1.663E C 09 2.443E C 08 0765 47.570 15965 0.446 1032 0.000 The first establish a regression model is as follows. MVA D C NI C MVA D C ROE C MVA D C EPS C MVA D C CFOPS C Put EVA into the regression equation above all, and set up the bivariate regression model, then review the explain ination increment. MVA D C NI C EVA C MVA D C ROE C EVA C MVA D C EPS C EVA C MVA D C CFOPS C EVA C By comparing statistical results of Tables 3 and 4, the ability of the traditional perance indicators NI,ROE, EPS, CFOPS of explanation of MVA get greatly Study on the Perance uation of Manufacturing Enterprises Based on EVA 691 Constant 6.079EC08 3.008 0.001 NI 82.465 14.130 0.000 8.652 EPS 1.809EC09 6.514 0.000 1.378 Table 5 Traditional index coefficient multivariate regression model Model i t-value Sig. VIF Constant 1.319EC09 3261 0.002 NI 52.404 20612 0.000 2.109 ROE 4.549EC09 0774 0.441 2.057 EPS 1.499EC09 4232 0.000 1.700 CFOPS 3.931EC08 1576 0.119 1.511 Table 6 After putting variables EVA into traditional index multiple regression model coefficient Model i t-value Sig. VIF EVA 33.771 5.598 0.000 7.385 increased after put variable EVA into the model, adjusted R2 increased by 31.32 , 87.34 , 1,472.34 and 2,644.44 respectively, shows that EVA provides incre- mental ination, even for some model, provides most of the explain ination, namely, EVA has contain some of the traditional perance indicators have no ination which explains the traditional perance indicators of MVA cant explain. Based on previous theoretical analysis, this part of ination refers to the equity capital cost, namely, and part of the impact of accounting ination distortion. The incremental ination is testified in a certain extent this hypothesis 2 that EVA is superior to the traditional perance uation index. In order to further verify this hypothesis 2, this paper constructs multiple regression model which the traditional perance indicators and EVA indicators as independent variable and the MVA as dependent variable. MVA D C 1NI C 2ROE C 3EPS C 4CFOPS C Put EVA into the regression equation above all, and set up the multivariate regression model, then review the explain ination increment. From the statistical results of Tables 5 and 6, the integral model of F value coefficient of each variable and model t-value are passed the significance test of significance level of 0.001. Contrast Tables 4 and 5 show that the model is adjusted to join the new variable EVA, adjusted model of R2 value increased from 89.2 to 92.8 . And the goodness of fit increases to a certain extent, which shows that after joining EVA, the capacity of model to explain the MVA increased. Observed variance inflation factor VIF is based values can be seen that after introducing variable EVA into the model, the degree of multicollinearity enhance. The reason is that EVA has the high correlation with NI. But due to the purpose of this study is not the accuracy of the regression coefficients of the variables, the overall perance analysis model of index explanation for market value MVA capacity, and considering the VIF is less than 10, do not belong to the significan