IMPACT OF CORPORATE GOVERNANCE ON FINANCIAL PERFORMANCE

Transcription

1 IMPACT OF CORPORATE GOVERNANCE ON FINANCIAL PERFORMANCE In this chapter, an attempt has been made to analyze the impact of corporate governance disclosure practices as per clause 49 of the listing agreement on financial performance of companies through regression analysis. A set of four firm performance measures have been selected, categorized into accounting based measures and market based measures to ascertain the relationship between corporate governance and firm performance. These measures are (i) Net Profit Margin on Sales, (ii) Return on Assets (ROA), (iii) Return on Equity (ROE) and (iv) Tobin s Q. First three measures are termed as accounting based measures and the fourth one is known as a measure of market valuation. The variables used for measuring the financial performance have been explained in section 7.1.The explanation of control variables used in the present study has been given in section 7.2. Section 7.3 explains the regression model employed for measuring the impact of corporate governance on financial performance of the companies under study and sections 7.4 and 7.5 give the results and discussions. 7.1 Measures of Financial Performance As explained above, four financial performance measures have been selected for the present study namely net profit margin on sales, return on assets (ROA), return on equity (ROE) and Tobin s Q. These are explained as given below: 7.1 (a) Net Profit Margin on Sales Net profit margin on sales ratio establishes the relationship between net profit and sales and indicates management s efficiency in manufacturing, administering and selling the products (Pandey, 2000, p.132). Higher the net profit margin better will be the profitability position of the company. Rechner et al. (1991), Dalton et al. (1999), Brown and Caylor (2004) and Phani et al. (2005) have used net profit margin on sales as a measure of firm performance in their studies. Net profit margin on sales has been calculated by dividing the earnings before interest and taxes (net of non-recurring

2 transactions) 1 by the net sales. Net sales are the sales excluding indirect taxes and duties such as excise duty and octroi duty. Sales are also net of internal transfers. Net Profit Margin on Sales= 7.1 (b) Return on Assets (ROA) EBIT Net Sales 100 ROA is used as an accounting based measure of firm performance. ROA is commonly used and well understood measure of firm performance, particularly appropriate for manufacturing firms (Kim, 2005). ROA measures the ability of the management to earn a return on resources and the firms using their assets efficiently have higher returns (Sharan, 2005, p.283). Various researchers namely Rechner et al. (1991), Klein (1998), Core et al. (1999), Dalton et al. (1999), Jog and Dutta (2004) and Phani et al. (2005) have used ROA as a firm performance measure in their studies. We have calculated return on assets by dividing the earnings before interest and taxes (net of nonrecurring transactions) by the total assets. Taxes are not controllable by the management and also one may not know the marginal corporate tax rate while analyzing the publishing data (Pandey, 2000, p.136). So, in order to remove this anomaly, we have considered EBIT instead of profit after tax (PAT). ROA = EBIT Total Assets (c) Return on Equity (ROE) Return on equity has been considered as another measure of firm performance. A number of researchers have employed ROE as firm performance measure in their studies (Rechner et al. 1991, Dalton et al, 1999, Rhoades et al., 2001, Brown and Caylor, 2004 and Jog and Dutta, 2004). ROE is an important indicator which tells us how the company has used the resources of its owners. This ratio reflects the extent to which the objective of wealth maximization of shareholders has been achieved. In the present study, we 1 As per Prowess Database of CMIE, net of non-recurring transactions includes profit or loss on sale of fixed assets and investments, provision written back, prior period income or expenses, insurance claims, etc. So, the above figure of EBIT is adjusted of NNRT. 231

3 calculated ROE by dividing the profit after tax (net of non-recurring transactions) adjusted with preference dividend by net worth minus preference share capital. 7.1 (d) Tobin s Q ROE= PAT Preference Dividend Net Worth-Preference Share Capital 100 Tobin s Q as a measure of market valuation has been extensively used by the researchers in their studies (Farrer and Ramsay, 1998, Chen, 2001, Mohanty, 2002, Weir et al., 2003, Brown and Caylor, 2004, Jog and Dutta, 2004, Dwivedi and Jain, 2005 and Khiari et al., 2005). Tobin s Q ratio has been devised by James Tobin. This ratio is based on the notion that combined market value of all the companies on the stock market should be equal to their replacement costs ( This is the ratio of market value of equity and debt divided by the replacement costs of total assets. Firms displaying Tobin s Q greater than unity are considered to be using scarce resources effectively, while those with Tobin s Q less than unity are using resources poorly (Chen, 2001). Tobin s Q as a measure of firm performance represents the value that investors put on in firm s shares above the total value of assets of the firm and thus represents investor s confidence which in turn is an indicator of the effectiveness of corporate governance mechanisms of the firm (Dwivedi and Jain, 2005). We calculated Tobin s Q ratio as a market value of equity plus book value of debt divided by book value of total assets. Since debt is not traded in the Indian stock market and replacement cost of assets is not available in case of Indian companies, so we used book value of debt and total assets. Hence, we computed it as follows: Tobin s Q = 365 Days Average Market Capitalization + Book Value of Debt Book Value of Total Assets 100 Debt here includes both short term and long term liabilities. So far as market capitalization is concerned, we have considered annual average market capitalization and this figure has been extracted out from the Prowess Database. 232

4 7.2 Control Variables The study of existing literature reveals that there are certain other factors also which may affect the performance of the firms. These factors are size of the firm, age, risk, leverage, industry type, etc. Therefore, it is essential to control these variables while analyzing the impact of corporate governance on financial performance. The details of control variables are given as below: 7.2 (a) Age Age is considered to be a significant factor affecting the firm performance. Due to the effects of learning curve and survival bias, older firms are considered to be more efficient than younger ones (Chen, 2001). Older firms have established themselves firmly in the market and are able to reap the benefits of the economies of scale which the younger ones or newcomers find it difficult to achieve. The researchers have controlled the impact of age while analyzing the impact of corporate governance mechanisms on firm performance (Chen, 2001, Mohanty, 2002, Jog and Dutta 2004, Kim 2005, Phani et al. 2005, Sheu and Yang 2005 and Mayur and Saravanan, 2006). 7.2 (b) Size of the Firm Numerous researchers have examined the relationship between size and performance of the firms. In product market, size reflects entry barriers that might result from economies of scale and in capital market, size reflects financial barrier of entry due to the ability of large companies to finance investment projects from internal sources as well from issue of new equity (Phani et al., 2005). There are mixed evidences available in the existing literature on relationship between size and firm performance. Chen (2001), Mohanty (2002), Weir et al. (2003), Mollah and Talukdar (2007) found out significant negative relationship between size and firm performance. On the other hand, Jog and Dutta (2004) and Kim (2005) found significant positive relationship between firm size and performance. There are no. of ways available for measuring the size of the firm. Chen (2001), Kim (2005), Wan and Ong (2005), Saravanan (2006) and Mollah and Talukdar (2007) have measured the size in terms of log of assets. While on the other hand, Fuerest and Kang (2000), Carson (2002), Mohanty (2002) and Jog and Dutta (2004) have considered market capitalization as proxy of a firm size. A few researches 233

5 have measured size in terms of sales also (Weir et al. 2003, Phani et al and Subramanian, 2006). In the present research, log of net sales has been used as a proxy of firm size. As there is curvilinear relationship between the size and firm performance, we have employed log of net sales instead of simply taking net sales figures. It is expected that firms with larger size outperform the smaller ones due to the certain advantages of economies of sales. So, it is essential to control the impact of size on firm performance. 7.2 (c) Risk Risk denotes some degree of hazard which can result on account of various factors such as short term fluctuations in profits, change in consumer tastes, change in technology, change in government policy, strategic moves of competitors, etc. Risk is associated with the future events. Since future is always uncertain and can t be predicted with accuracy. So, entrepreneurs have to take future decisions by keeping in mind the risk factor. Based on the concept of higher the risk, higher will be the return, various researchers have tried to find out the relationship between risk and profitability of the firm. In line with Mohanty (2002), Jog and Dutta (2004) and Mollah and Talukdar (2007) beta has been incorporated as a measure of risk. 7.2 (d) Leverage Leverage has also been employed by researchers in their studies on firm performance (Daily and Dalton, 1994; Chen, 2001; Carson, 2002; Dwievedi and Jain, 2005; Khiari, et al. 2005; Kim, 2005 and Mayur and Saravanan, 2006). When the firm s cost of debt is lower than the firm s rate of return on its assets, then shareholders returns in form of EPS and return on equity increase and hence, leverage will have favourable impact on profitability. However, shareholders returns will fall, when the firm obtains the debt at higher cost than the rate of return on its assets. There are mixed evidences available in the literature on the relationship between leverage and profitability. Chen (2001) found out the negative relationship between leverage and profitability. On the other hand, Kim (2005), Khiari et al. (2005) examined the positive association between leverage and profitability. There are various measures of financial leverage employed by the researchers. But in line with Chen, 2001, Khiari et al., 2005 and Kim 2005 debt- 234

6 equity ratio has been employed as a measure of financial leverage. We have used the following formula to compute debt-equity ratio of the firm. Leverage = Long Term Debt Shareholder s Funds (e) Industry Effects Industry characteristics have vital concern in the analysis of firm performance. Firms in new and expanding industries are expected to outperform those operating in old and declining industries (Kumar, 1985, Singh, 1997 and Kaur, 2005). The firms in those industries, where exist growth opportunities, concentrated competitors and stable markets should have higher profits than industries that are in decline (Coles et al., 2001). It has been persistently shown that firms in a particular industry earn comparatively above normal profits by virtue of some favourable structural characteristics (Amato and Wilder, 1990). Like Mohanty (2002), Dwivedi and Jain (2005) and Mollah and Talukdar (2007), we have captured the industry effects by introducing 8 industries dummies in regression model. Table: 7.1 Summary of Control Variables Used in Various Studies S.No Variables Size Age Risk Leverage Industry Effects Studies 1 Daily and Dalton (1994) 2 Klein (1998) 3 Fuerest and Kang (2000) 4 Chen (2001) 5 Carson (2002) 6 Mohanty (2002) 7 Weir et al. (2003) 8 Jog and Dutta (2004) 9 Dwivedi and Jain (2005) 10 Khiari et al. (2005) 11 Kim (2005) 12 Phani et al. (2005) 13 Sheu and Yang (2005) 14 Wan and Ong (2005) 15 Mayur and Saravanan (2006) 16 Subramanian (2006) 17 Mollah and Talukdar (2007) 235

7 7.3 Regression Model In order to find out the impact of corporate governance on financial performance of the companies, ordinary least square (OLS) regression model with enter method has been employed. SPSS version has been used to compute the results of multiple regression models. The following models have been used to analyze the impact of corporate governance on various measures of financial performance: Model I Net Profit Margin on Sales = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) + 7 (D_Iron & Steel) + 8 Model II (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + ε Return on Assets = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) + 7 (D_Iron & Steel) + 8 Model III (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + ε Return on Equity = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) Model IV + 7 (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + ε Tobin s Q = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) 236

8 Whereas (D_Industry) represents industry dummies. + 7 (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + ε In the above four models, Net Profit Margin on Sales ratio, ROA ratio, ROE ratio and Tobin s Q ratio have been used as measures of firm s performance and considered as dependent variables. However, corporate governance, firm characteristics and industry dummies have been used as explanatory variables. β 0 is a constant term and ε denotes error term here. 1, 2, are regression coefficients. One industry dummy has been omitted from the above four models. Dummy variables are essentially a device to classify the data into mutually exclusive categories (Gujarati, 2004, p.298). If the variable (here industry) has m categories then we include m-1 dummy variables. In our analysis, we have the sample of 9 industries and 8 dummies have been introduced. The industry for which no dummy variable has been assigned is known as base or reference industry. In the present analysis, power industry has been omitted and hence, known as base industry. The coefficients attached to the dummy variables are known as the differential intercept coefficients as they tell how much value of the intercept that receives the value of 1 differs from the intercept coefficient of the benchmark category (Gujarati, 2004, p.302). We assigned code 1 to all the firms who are members of a particular industry category and 0 for otherwise. The above stated models have been run with both unequal weights and equal weights methods. Hence, 8 regression models have been obtained with same dependent variable, control variables and industry dummies. All the listed companies were required to adopt the provisions of clause 49 of the listing agreement up to the year So, there exists difference in the number of companies for the first three years of analysis. For the last two years, difference in the number of companies 237

9 is due to the missing data. So, in order to overcome this problem, regression models have been employed year wise instead of taking average of the whole data. While applying the regression models the assumptions of normality, multicollinearity and autocorrelation have been tested. The assumption of normality of data has been checked out through descriptive statistics i.e. skewness and kurtosis. With the help of SPSS package, both standard error and statistics value have been computed. Z value has been derived manually for all the variables by dividing the standard error with statistic value. If calculated value of z exceeds ± 2.58 and ± 1.96 then we can reject the assumption of normality of data at.01 and.05 significance levels respectively. Normality has been checked out for age, risk and net sales. From the existing literature, no evidence has been found out for checking the normality of leverage. So, in line with the existing research evidence, leverage has been excluded from the test of normality. Age and risk found to be at normal distribution level. So far as net sales were concerned, test statistics have rejected the assumption of normality. After checking out the normality of the data, the next step is to transform the variable which is not normal. There are various measures suggested by the researchers for transformation of data such as taking square root of the variable, logarithms and inverse of the variable. However, net sales have been transformed by taking its logarithm. The presence of multicollinearity among the independent variables may affect the overall regression results and will lead to wrong estimations. In order to detect the problem of multicollinearity, variance inflation factor (VIF) has been computed for each of the individual independent variables by using SPSS package. As a rule of thumb, if VIF of variable exceeds 10, then there is a problem of multicollinearity with that variable. In line with the above rule of thumb, we did not find any variable in our analysis, whose VIF value exceeds 10. Hence, our data is free from multicollinearity problem. 238

10 The next assumption of autocorrelation of multivariate analysis has been checked through Durbin Watson (d) statistic value. As per decision rule, if d is 2 or close to 2, then there is no first order autocorrelation either positive or negative. In our analysis, data is free from autocorrelation problem also. 7.4 Results and Discussions This section explains the regression results of all the four models stated in Sec 7.3. Table 7.2(a) summarizes the regression results of impact of corporate governance with equal weights on net profit margin on sales. F statistics are significant at 1% level for all the five years which reveal that our model is appropriate. Corporate governance is found to be significantly (5% level) positively associated with net profit margin on sales for just one year out of the period of five years. However, adjusted R² is also high for this year i.e One of the possible reasons for this phenomenon might be the adoption of more no. of items of clause 49 of listing agreement by the companies in the second year of its implementation. For the years and , negative but insignificant relationship has been found out between corporate governance and net profit margin on sales. Similarly, for the years and , though there is positive relationship between corporate governance and firm performance yet the relationship is insignificant one. Age is found to be negatively associated with net profit margin on sales for the period of five years but it is found to be significant for the years , and only. There are mixed evidences available in the literature regarding the relationship between age and firm performance. However, our findings are in line with Chen (2001), Singh (1997), Sheu and Yang (2005) and Kaur (2005). One of the possible reasons for such a negative relationship could be the stickiness of older firms towards older technology and old means of production. These firms resist for the adoption of new technology available these days. Moreover, as per the product life cycle theory, firms generally earn higher profits during the growth and maturity stage. But as soon as the maturity phase gets over and declining stage starts, profits start declining. Thus, age factor after a certain period of time has negative impact on profitability. 239

11 Size is found to have positive and significant relationship with net profit margin on sales for four out of five years. Our findings in this context are in consonance with Jog and Dutta, 2004 and Kim, 2005 and hold the view that firms with larger size outperform the smaller ones in size because of certain advantages of economies of scale. Risk has negative but insignificant relationship with net profit margin on sales for all the five years. In other words, no significant impact of risk has been found out on net profit margin on sales. Our findings in this context do not support the notation of Higher the risk, higher will be the return. One of the possible reasons of negative relationship between risk and profitability can be attributable to risk management strategies adopted by the firm where managers always seek to minimize the risk and maximize the profits. Leverage is found to be significantly negatively associated with net profit margin on sales for the year only. However, the nature of relationship is not found to be consistent for the whole period under study. For the first two years, a positive but insignificant relationship has been observed between leverage and net profit margin on sales. Afterwards, this relationship becomes negative for the rest of the three years period. One of the reasons of this negative relationship can be explained in terms of cost of capital. Due to increase in debt- equity ratio of the firm, the cost of capital will increase which will ultimately lower the profitability. So far as sectoral effects are concerned, it has been observed that textiles, iron and steel, automobile, cement and sugar industries are showing relationship with financial performance for a few years only. Textile industry has highly significant but negative relationship with net profit margin on sales for four out of five years. Iron and steel industry has also been showing consistently negative relationship with net profit margin on sales from to However, this relationship is found to have significant association for first four years only. Automobile industry has been significantly negatively associated with net profit margin on sales for the entire period of study. Similarly, cement industry is found to have negative association with net 240

12 profit margin on sales but this relationship is significant for the period of four years i.e. from to Drugs & Pharmaceuticals and paper industries have negative but insignificant association with net profit margin on sales. Software industry has insignificant positive association with net profit margin on sales. Sugar industry is found to have negative relationship with net profit margin on sales for the entire period of study but the relationship is found to be significant for three years only i.e. from to So far as regression results with unequal weights of corporate governance are concerned, no significant variations in R² and adjusted R² values have been observed. With both the methods the results have been found to be almost same for all the explanatory variables. However at some places, the level of significance is varying. Table 7.2(b) reveals that corporate governance is significantly (10% level) associated with net profit margin on sales for the year For the rest of the years, no significant impact has been found for corporate governance on net profit margin on sales. Similarly, age is found to be significantly negatively associated at 5% level with net profit margin on sales for the years , and Size is found to be highly significant but positively associated with net profit margin on sales for four out of five years. It has been observed from the results that there is no impact of risk on net profit margin on sales. Mixed results have been found out for association of leverage with financial performance of the companies. There is negative but significant association of leverage with net profit margin on sales for the years and So far as sectoral effects are concerned, it has been found out that automobile industry has significant negative association with the net profit margin on sales. Textiles, iron & steel and cement industries have been found to be significantly negatively associated with net profit margin on sales for four out of five years, whereas drugs & pharmaceuticals and paper industries have negative but insignificant relationship with net profit margin on sales. On the other hand, no significant relationship has been observed of software industry with net profit margin on sales. Sugar industry is also found to have significant negative association with net profit margin on sales for three years out of five years. 241

13 Table: 7.2 (a) Regression Results of Impact of Corporate Governance (Equal Weights) on Net Profit Margin on Sales S.No Explanatory Variables Corporate Governance (-.10).183** (2.02).142 (1.45).015 (.163) (-.594) 2 Age (-.67) -.27* (-2.79) -.26** (-2.54) -.188** (-2.056) (-.516) 3 Size.191 (1.49).232** (2.59).264* (2.67).381* (4.126).287* (2.781) 4 Risk (-.30) (-.58) (-.037) (-.883) (-1.472) 5 Leverage.103 (.87).071 (.820) -.15 (-1.58) -.213** (-2.478) (-1.590) 6 Textiles -.26 (-1.41) -.45* (-3.86) -.32** (-2.54) -.386* (-3.308) -.420* (-3.432) 7 Iron & Steel -.32*** (-1.89) -.49* (-4.3) -.26** (-2.13) -.223** (-2.026) (-.826) 8 Automobile -.48** (-2.6) -.53* (-4.37) -.44* (-3.31) -.553* (-4.557) -.607* (-4.789) 9 Cement -.32 (-1.58) -.31** (-2.59) -.29** (-2.34) -.256** (-2.257) -.253** (-2.136) 10 Drugs & Pharmaceuticals (-.35) -.14 (-1.14) (-.56) (-.465) (-1.249) 11 Software.37 (1.53).12 (.91).060 (.39).010 (.074).081 (.575) 12 Sugar -.19 (-1.09) -.24** (-1.96) -.24*** (-1.83) -.244** (-2.025) (-1.014) 13 Paper (.33) -.16 (-1.29) (-.66) (-.999) (-1.596) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 242

14 Table: 7.2 (b) Regression Results of Impact of Corporate Governance (Unequal Weights) on Net Profit Margin on Sales S.No Explanatory Variables Corporate Governance (-.076).153*** (1.661).146 (1.481).027 (.309) (-.649) 2 Age (-.686) -.259** (-2.650) -.265** (-2.578) -.192** (-2.102) (-.500) 3 Size.189 (1.511).248* (2.746).271* (2.771).378* (4.122).288* (2.812) 4 Risk (-.302) (-.692) (-.026) (-.897) (-1.428) 5 Leverage.103 (.868).071 (.822) -.156*** (-1.653) -.215** (-2.495) (-1.565) 6 Textiles (-1.408) -.447* (-3.771) -.331** (-2.573) -.388* (-3.329) -.419* (-3.419) 7 Iron & Steel -.318*** (-1.877) -.479* (-4.200) -.252** (-2.095) -.223** (-2.035) (-.846) 8 Automobile -.477** (-2.600) -.520* (-4.242) -.443* (-3.326) -.556* (-4.604) -.607* (-4.797) 9 Cement (-1.580) -.297** (-2.498) -.299** (-2.372) -.258** (-2.279) -.253** (-2.141) 10 Drugs & Pharmaceuticals (-.346) (-1.020) (-.606) (-.489) (-1.244) 11 Software.367 (1.536).156 (1.157).069 (.466).008 (.059).076 (.544) 12 Sugar (-1.090) -.234*** (-1.878) -.241*** (-1.855) -.246** (-2.040) (-1.032) 13 Paper (-.319) (-1.207) (-.624) (-1.003) (-1.628) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 243

15 Table 7.3 (a) reveals the regression results of impact of corporate governance with equal weights on return on assets. F-statistics are highly significant at 1% level for all the five years which show that our model is fit to be used. It has been observed that corporate governance is significantly positively associated with the return on assets for two out of five years. For the years and , there is positive but insignificant relationship between corporate governance and return on assets. Age is found to be significantly negatively associated with the return on assets for two out of five years. Size is found to have highly significant but positive relationship with return on assets for the entire period of five years. So far as risk is concerned, there exists negative but significant relationship with return on assets from the year to Leverage is found to have negative but significant relationship with return on assets for the year only. It has been observed from the industry effects that software industry has significant positive impact on return on assets. Textiles industry is found to have significant negative relationship with ROA for one out of five years only. On the other hand, iron & steel industry is found to be significantly associated though negatively with return on assets for the year This relationship turns positive and significant in the year Drugs & Pharmaceutical industry has positive but significant relationship with profitability for three out of five years. Sugar industry has significant positive relationship with return on assets in the year only. For the rest of the industries dummies, no significant relationship has been observed with ROA. Similarly, table 7.3 (b) reveals almost the similar results as reported in table 7.3 (a). The impact of corporate governance has been observed on ROA of the companies for two years i.e and So far as other explanatory variables are concerned, we didn t find much variation in their β coefficients. The value of F statistics is also significant at 1% level for the entire period of five years. Similarly, the values of adjusted R² are also almost same as given in table 7.3 (a). 244

16 Table: 7.3 (a) Regression Results of Impact of Corporate Governance (Equal Weights) on Return on Assets S.No Explanatory Variables Corporate Governance (-.77).186** (2.13).20** (2.04).041 (.428).005 (.048) 2 Age (-.68) -.22** (-2.37) -.21** (-2.12) (-1.257).035 (.361) 3 Size.32* (2.7).26* (2.99).36* (3.66).382* (3.911).341* (3.270) 4 Risk -.37** (-2.53) -.31* (-2.99) -.24** (-2.19) -.234** (-2.102) -.352* (-3.192) 5 Leverage.039 (.36).044 (.53) -.11 (-1.18) -.221** (-2.425) (-1.075) 6 Textiles.027 (.16) -.22** (-1.95) (-.76) (-.727) (-.617) 7 Iron & Steel (-.62) -.25** (-2.32) (-.44).109 (.934).331* (2.843) 8 Automobile (-.49) -.12 (-1.05).006 (.044) (-.082) (-.881) 9 Cement -.13 (-.71) -.14 (-1.22) -.14 (-1.16) (-.198).049 (.407) 10 Drugs & Pharmaceuticals.24 (1.27).21*** (1.78).29** (2.26).324** (2.567).142 (1.128) 11 Software.90* (4.06).50* (3.79).25*** (1.71).269*** (1.90).445* (3.126) 12 Sugar.004 (.021) (-.13) (-.06).035 (.271).241*** (1.898) 13 Paper.089 (.56).018 (.15).095 (.72).106 (.934).076 (.583) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 245

17 Table: 7.3 (b) Regression Results of Impact of Corporate Governance (Unequal Weights) on Return on Assets S.No Explanatory Variables Corporate Governance (-1.198).159*** (1.787).195** (2.016).031 (.336) (-.375) 2 Age (-.579) -.213** (-2.260) -.219** (-2.164) (-1.227).046 (.467) 3 Size.319* (2.788).276* (3.169).367* (3.806).386* (3.977).359* (3.462) 4 Risk -.375** (-2.603) -.309* (-3.047) -.254** (-2.268) -.235** (-2.105) -.346* (-3.122) 5 Leverage.037 (.344).045 (.531) (-1.271) -.223* (-2.445) (-1.073) 6 Textiles.023 (.138) -.213** (-1.863) (-.797) (-.714) (-.585) 7 Iron & Steel (-.549) -.246** (-2.237) (-.358).111 (.962).329* (2.835) 8 Automobile (-.410) (-.942).004 (.031) (-.055) (-.838) 9 Cement (-.695) (-1.144) (-1.195) (-.181).051 (.428) 10 Drugs & Pharmaceuticals.247 (1.297).228*** (1.919).285** (2.188).327** (2.600).149 (1.184) 11 Software.919* (4.199).525* (4.043).268*** (1.842).275** (1.978).451* (3.201) 12 Sugar.003 (.016) (-.034) (-.095).037 (.288).244*** (1.930) 13 Paper.089 (.563).029 (.242).103 (.790).110 (.858).079 (.610) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 246

18 Table 7.4 (a) reveals the regression results of impact of corporate governance on return on equity. No significant relationship has been observed between corporate governance and return on equity for the entire period of study. Age is found to be significantly negatively associated with the firm performance for one out of five years. Size is significantly positively associated with return on equity for the year only. So far as risk is concerned, a significant negative association has been observed with return on equity for the first two years only. Similarly, leverage is highly significant but negatively related to return on equity from the year to However, from the industry effects, it has been observed that that iron & steel industry has significant negative association with return on equity for the years and whereas this relationship turns positive for the year Software and paper industries have significant positive association with firm performance for one out of five years. For rest of the industries dummies, no significant relationship has been observed with return on equity. The value of adjusted R² has been observed to be at maximum level for the year Similarly, table 7.4 (b) reveals the regression results of impact of corporate governance on return on equity with unequal weights of corporate governance. No significant relationship has been observed between corporate governance and return on equity for the entire period of study. Age is found to be significantly negatively associated with the firm performance for the year only. In the same way, size has positive but significant impact on return on equity for the year only. Risk is found to be significantly negatively associated with return on equity for first two years only. So far as leverage is concerned, a highly significant but negative association has been observed with return on equity for the period of four out of five years. Iron & steel industry has significant negative association with return on equity for the years and but this relationship turns positive for the years and Paper industry has significant positive relationship with the return on equity for the year only. For rest of the industry dummies, no significant relationship has been observed with return on equity. 247

19 Table: 7.4 (a) Regression Results of Impact of Corporate Governance (Equal Weights) on Return on Equity S.No Explanatory Variables Corporate Governance.172 (1.27).092 (.95).036 (.41) (-.179) (-.089) 2 Age.035 (.25) (-.28) -.13 (-1.44) -.155** (-2.182) (-.176) 3 Size (-.041).10 (1.05).061 (.69).248* (3.454) (-.309) 4 Risk -.42** (-2.47) -.44* (-3.89).035 (.35) (-.160) (-.431) 5 Leverage.056 (.43) -.31* (-3.36) -.76* (-8.99) -.783* ( ) -.710* (9.240) 6 Textiles.033 (.16) (-.98) (-.18).031 (.348).010 (.093) 7 Iron & Steel -.47** (-2.54) -.39* (-2.99).17 (1.61) (-.095).165*** (1.688) 8 Automobile -.15 (-.76) -.15 (-1.16) (-.39) (-.728).032 (.298) 9 Cement -.19 (-.88) -.13 (-1.02).075 (.67).069 (.782).052 (.514) 10 Drugs & Pharmaceuticals.054 (.24).004 (.029) (-.28).133 (1.436).031 (.290) 11 Software.39 (1.49).30** (2.07) -.12 (-.91).014 (.131).014 (.118) 12 Sugar (-.45) (-.42).06 (.54).141 (1.502).161 (1.511) 13 Paper (-.11) (-.18) (-.77).313* (3.306) (-.593) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 248

20 Table: 7.4 (b) Regression Results of Impact of Corporate Governance (Unequal Weights) on Return on Equity S.No Explanatory Variables Corporate Governance.113 (.875).026 (.271) (-.213).014 (.200) (-.574) 2 Age.054 (.386) (-.026) (-1.211) -.163** (-2.300) (-.054) 3 Size.020 (.145).107 (1.118).074 (.848).240* (3.367) (-.116) 4 Risk -.421** (-2.451) -.431* (-3.881).038 (.376) (-.179) (-.360) 5 Leverage.057 (.439) -.310* (-3.362) -.757* (-9.022) -.784* ( ) -.710* (-9.250) 6 Textiles.033 (.164) (-.864) (-.054).027 (.297).014 (.131) 7 Iron & Steel -.472** (-2.547) -.339* (-2.804).185*** (1.726) (-.126).163*** (1.675) 8 Automobile (-.762) (-.924) (-.209) (-.807).037 (.348) 9 Cement (-.880) (-.842).091 (.809).064 (.729).054 (.537) 10 Drugs & Pharmaceuticals.049 (.214).029 (.224) (-.092).127 (1.372) ) 11 Software.423 (1.616).326 (2.287) (-.729).005 (.049).019 (.159) 12 Sugar (-.473) (-.320).074 (.640).137 (1.467).164 (1.542) 13 Paper (-.201) (-.046) (-.674).310* (3.292) (-.576) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 249

21 Table 7.5 (a) reveals the regression results of impact of corporate governance with equal weights on Tobin s Q ratio. Adjusted R² is high for the year and explains 34% of variation in model. F- statistics are significant at 1% and 5% level which depicts that our model is appropriate for the entire period of study. It has been observed from the β coefficients of corporate governance score that there is no significant relationship between corporate governance and Tobin s Q. Our findings are in consonance with Jog and Dutta (2004) who examined no significant relationship between corporate governance variables and firm performance measured by Tobin s Q. Age is found to be significantly positively associated with firm performance for three out of five years. Similarly, size is positively and significantly associated with firm performance for the period of two out of five years. Risk is found to be negatively but significantly related to Tobin s Q for the years and whereas no significant association has been observed between leverage and Tobin s Q. Textiles industry has been found to be significantly positively related to Tobin s Q for the period of four out of five years. Cement industry is found out to be significantly positively related to firm performance for the year only. For drugs and pharmaceutical industry, a significant positive association has been observed with Tobin s Q for the period of two out of five years. So far as software industry is concerned, a significant but positive association has been observed with Tobin s Q for the entire period of study. For rest of the industries dummies, no significant association has been observed with Tobin s Q. Table 7.5 (b) reveals the regression results with unequal weights of corporate governance. It has been observed from the analysis that there is no significant impact of corporate governance on firm performance as measured by Tobin s Q. For rest of the variables, the regression results are almost same as depicted in table 7.5 (a) except for drugs and pharmaceuticals and cement industries. For cement industry, results are found to be significant at 11% level for the year as compared to 10% reported in table 7.5 (a). So far as drugs and pharmaceuticals industry is concerned, a significant but positive association has been observed for three out of five years. F- statistics in this model are significant at 1% and 5% level which reveals that our model is fit for the entire period of study. Adjusted R² is the highest for the year and explains 34% variation in model. 250

22 Table: 7.5 (a) Regression Results of Impact of Corporate Governance (Equal Weights) on Tobin s Q S.No Explanatory Variables Corporate Governance (-.61).076 (.701).077 (.70) (-.062) (-.754) 2 Age.43* (3.09).264** (2.31).28* (2.48) (-.854) (-.099) 3 Size.068 (.49).068 (.63).063 (.57).319* (3.266).383* (3.779) 4 Risk -.37** (-2.18) -.12 (-.91) -.14 (-1.15) -.236** (-2.116) (-1.566) 5 Leverage (-.25).086 (.83).054 (.51).020 (.220).038 (.426) 6 Textiles.40** (2.04).31** (2.22).37** (2.59).205*** (1.666).090 (.749) 7 Iron & Steel.14 (.79).013 (.099).038 (.28).005 (.041) (-.300) 8 Automobile.148 (.752).032 (.22).057 (.38).088 (.687).089 (.716) 9 Cement.17 (.78).047 (.33).059 (.42).165 (1.379).192*** (1.649) 10 Drugs & Pharmaceuticals.25 (1.13).22 (1.51).18 (1.24).437* (3.471).468* (3.822) 11 Software 1.05* (4.07).49* (2.97).47* (2.81).612* (4.336).607* (4.392) 12 Sugar.074 (.38) (-.085) (-.12).009 (.068).060 (.485) 13 Paper.021 (.11) (-.15) (-.13).064 (.500).065 (.517) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 251

23 Table: 7.5 (b) Regression Results of Impact of Corporate Governance (Unequal Weights) on Tobin s Q S.No Explanatory Variables Corporate Governance (-1.146) (-.013).010 (.092).002 (.025) (-.357) 2 Age.445* (3.288).296** (2.529).302** (2.627) (-.855) (-.195) 3 Size.074 (.556).072 (.663).080 (.726).313* (3.224).365* (3.620) 4 Risk -.379** (-2.261) (-.834) (-1.124) -.237** (-2.119) (-1.570) 5 Leverage (-.273).083 (.797).050 (.474).016 (.176).040 (.451) 6 Textiles.394** (2.029).330** (2.323).387* (2.684).204*** (1.651).087 (.725) 7 Iron & Steel.158 (.878).036 (.264).054 (.401).005 (.047) (-.313) 8 Automobile.165 (.842).067 (.457).083 (.553).086 (.672).083 (.666) 9 Cement.175 (.800).074 (.517).077 (.547).163 (1.365).189 (1.620) 10 Drugs & Pharmaceuticals.256 (1.154).252*** (1.710).207 (1.403).436* (3.467).462* (3.760) 11 Software 1.081* (4.244).518* (3.216).505* (3.054).611* (4.387).595* (4.330) 12 Sugar.073 (.378).001 (.008) (-.022).007 (.051).054 (.439) 13 Paper.017 (.092) (-.013) (-.024).072 (.561).058 (.462) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 252

24 7.4.1 Regression Results with Pooled Data In pooled data, the elements of both time series and cross section units are present. In the previous section, an attempt has been made to study the impact of corporate governance on financial performance of the companies on yearly basis. The corporate governance code was not applicable to all the listed companies in the first year of its implementation. Hence, we found inconsistency in the no. of observations for the first three years of study. Results were also not consistent for the entire period of study. In order to be more conclusive, an attempt has been made to separate those companies which were following the corporate governance practices as per clause 49 of the listing agreement for the period of five years under study. A total no. of 68 companies has been selected from the sample of 112 which started following the corporate governance code from the first year of its implementation. After pooling the data of 68 companies and applying the industries dummies as well as year dummies, the following regression models have been used: Model 1 Net Profit Margin on Sales = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) Model (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + 14 (D 1 ) + 15 (D 2 ) + 16 (D 3 ) + 17 (D 4 ) + ε Return on Assets = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) Model (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + 14 (D 1 ) + 15 (D 2 ) + 16 (D 3 ) + 17 (D 4 ) + ε Return on Equity = (Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) + 7 (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 253

25 Model 4 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + 14 (D 1 ) + 15 (D 2 ) + 16 (D 3 ) + 17 (D 4 ) + ε Tobin s Q = Corporate Governance) + 2 (Age) + 3 (Size) + 4 (Risk) + 5 (Leverage) + 6 (D_Textiles) + 7 (D_Iron & Steel) + 8 (D_Automobile) + 9 (D_Cement) + 10 (D_Drugs & Pharmaceuticals) + 11 (D_Software) + 12 (D_Sugar) + 13 (D_Paper) + 14 (D 1 ) + 15 (D 2 ) + 16 (D 3 ) + 17 (D 4 ) + ε Here, D1, D2, D3 and D4 represent year dummies Table 7.6 (a) represents the regression results of impact of corporate governance with equal weights on financial performance of the companies. Results revealed no significant association between corporate governance and financial performance. Age is found to be significantly positively associated with one of the financial performance measures i.e. Tobin s Q. It has been observed that size has positive and significant impact on financial performance. However, significant negative association has been observed between risk and financial performance. Leverage is found to have negative and significant impact on measures of profitability. But so far as measure of market valuation is concerned, no association has been observed with leverage. Regarding the sectoral effects, textiles industry is found to be significantly negatively associated to some extent with profitability but positive impact has been observed on the measure of market valuation. Iron and steel industry is found to be significantly negatively associated with net profit margin on sales and ROE. However, the same is significantly positively associated with ROA. So, mixed results have been observed for the measures of profitability. Automobile and cement industries are highly significant but negatively related to one of the measures of profitability i.e. net profit margin on sales. Hence, it is related to some extent with profitability. A positive impact has been observed for drugs & pharmaceuticals industry on one of the measures of profitability and market valuation i.e. ROA and Tobin s Q respectively. So far as software industry is concerned, results revealed highly significant but positive relationship with financial performance. Regarding the year dummies, we observed that year has significant positive impact on Tobin s Q. On the other hand, year is related negatively to financial performance in terms of ROE. For rest of the year dummies, no significant impact has been observed on financial performance. 254

26 Table: 7. 6 (a) Regression Results of Impact of Corporate Governance (Equal Weights) on Financial Performance S.No Explanatory Variables Model I Model II Model III Model IV 1 Corporate Governance (-.128).001 (.023).082 (1.282).029 (.443) 2 Age (-.025).010 (.186).091 (1.614).155* (2.740) 3 Size.194* (3.492).325* (5.879).111*** (1.867).192* (3.208) 4 Risk -.141** (-1.997) -.396* (-5.641) -.339* (-4.506) -.173** (-2.273) 5 Leverage -.112** (-2.090) -.138** (-2.597) -.144** (-2.521).003 (.048) 6 Textiles -.354* (-4.175) (.106).046 (.510).185** (2.039) 7 Iron & Steel -.203** (-2.530).151*** (1.896) -.195** (-2.286).039 (.459) 8 Automobile -.471* (-5.610).014 (.163) (-.254).063 (.703) 9 Cement -.313* (-3.370) (-.107) (-.567).114 (1.142) 10 Drugs & Pharmaceuticals (-1.229).292* (3.092).121 (1.199).255** (2.496) 11 Software.217** (2.110).684* (6.703).371* (3..397).675* (6.118) 12 Sugar -.237* (-2.882).076 (.933).017 (.191).062 (.703) 13 Paper (-1.588).073 (.916).029 (.335).054 (.629) 14 D (-.081).049 (.765) (-1.529).182* (2.615) 15 D (-.811).020 (.330) -.151** (-2.376).053 (.824) 16 D (-.950) (-.218) (-1.476).038 (.606) 17 D (-.952).005 (.080) (-.684) (-.178) R Square Adj. R Square F Statistics ( Significance) Note: Figures within the parentheses indicate the t-values, (*) indicates significance at 1% level, (**) indicates significance at 5% level and (***) indicates significance at 10% level. 255