CHAPTER 6 DATA ANALYSIS: CORPORATE DIVIDEND POLICY DETERMINANTS

Transcription

1 CHAPTER 6 DATA ANALYSIS: CORPORATE DIVIDEND POLICY DETERMINANTS The body of literature dealing with dividend determinants can be grouped into two distinct categories: (1) those based on the implicit assumption of symmetric information and (2) those based on the explicit assumption of asymmetric information. In the symmetric information models, the seminal work is that of Lintner (1956). According to Lintner s model, current dividends are predicated upon past dividends and current profits. Lintner s two-variable model is supported by the empirical evidence of Fama and Babiak (1968) 35. Theories based on the supposition of asymmetric information include agency, pecking order, and most profoundly, dividend-signaling theories. Agency theory explanations of dividend build on the works of Easterbrook (1984) and Jensen (1986). The pecking order theory (the complete opposite of the agency theory explanation) is espoused by Myers (1984) and Myers and Majluf (1984).[19],[20] Signaling theory, first proposed by Bhattacharya (1979, 1980), asserts that good firms are able to signal their expected fortunes via the disbursement of dividends. The tax costs of which are fully recovered by ensuing stock price increases.[[33],[34] According to Miller and Modigliani, investment decisions are independent of financing decisions such as new debt or dividends because, in perfect capital markets, the value of a firm is affected only by its investment decision. And both the financing decision and dividend policy are irrelevant in the process of determining the firm s value.[9] 35 A somewhat different work is that of Marsh and Merton (1987), who developed a dynamic aggregate dividend behavior model based on both accounting data and stock market data. In contrast to Marsh and Merton, the present study treats the dividend policy problem endogenous to the firm. In addition, there are marked differences between the two studies, both in method of analysis and in data used. Nevertheless, the conclusions of this study and the Marsh and Merton study concerning the validity of signaling models are surprisingly similar. 193

2 Dhrymes and Kurz (1967) and McCabe (1979) found that the firm s investment decision is linked to its financing decision. Higgins (1972), Fama (1974), and Smirlock and Marshall (1983) documented no interdependence between investments and dividends. It is argued here that these tests seem to be misspecified or inappropriate because they have omitted important explanatory variables and/or lagged variables. [92] The agency theory explanation of dividend payment is that the firm should pay dividends to reduce agency costs associated with excess cash flows. Easterbrook (1984) further argued that paying dividends can reduce agency costs because it forces managers to return to the market for cash, thus keeping them under constant monitoring. [19] In contrast, the pecking order hypothesis offers and Majluf (1984) states that dividend payments are negatively influenced by free cash flows and Majluf (1984) claim that firms can build up financial slack by restricting dividends when investment requirements are modest. Thus, Dividend decision in the corporate sector is governed by a large number of determinants. The foregoing discussions on the review of empirical studies disclose that profit, cash flow, lagged dividend, capital expenditure, retained earnings, flow of external fund, cost of debt, changes in sales, share price behaviour etc., are expected to have a direct bearing on the dividend policy decision of the firms. In the following segment these variables and their relationship with dividend payout has been discussed at length. 6.1 LEADING DETERMINANTS OF DIVIDEND POLICY Dividend decisions in the corporate sector are governed by large number of determinants. The various research studies done abroad and in Indian context have been detailed in the literature review chapter. Enumerated below are the key variables identified as per available literature along with the relationship with dividend payout ratio of the firm. 194

3 PROFITABILITY The key determinant of dividend payments is the current earnings which represents the capacity of a firm to pay dividends. Profitability has a positive relationship with dividends. Research studies have used PAT, ROE and RONW as proxies for profitability of the company. CASH FLOW Brittain (1966) suggested that cash flow (net current profit after tax + depreciation) is a more appropriate measure of the company s capacity to pay dividend. Further the regulations and accounting practices regarding depreciation allowance keep on changing and as such cash flow may be a better indicator of true earnings than net profit. LAGGED DIVIDEND The specification of dividend equation by Lintner (1956) suggested that lagged dividend is the only other explanatory variable of dividend policy (first being net profit). The rationale of lagged dividend as a determinant of dividend policy is provided by the speed of adjustment mechanism, which states that companies try to achieve a certain desired payout ratio in the long run. In order to follow a stable dividend policy management has to allow the past dividend trend to influence the current dividend payments. DEBT EQUITY RATIO The demand for external finance usually arises in a company on account of constraints imposed by its internal resources. Higher the internal flows, given the investment requirements, lesser will be the demand for borrowings and vice-versa. That is, higher the dividend, larger the demand for borrowings and higher is the debt equity ratio. Firms with high debt ratios ought to pay lower dividends as they have already precommitted their cash flows to make debt payments. Through lower dividend payout firm may avoid borrowing more capital. 195

4 SALES GROWTH An increase in sales generates increased working capital requirement, which in turn may adversely affect dividend payments. Some research studies have also taken sales growth as proxy for growth and investment opportunities available to firm. This implies a negative relationship between dividend and sales growth. LIQUIDITY A firm may have adequate earnings to declare dividends, but it may not have sufficient cash to pay the same. The liquidity position of a company is expected to be positively related to dividend payment. Current ratio and quick ratio has been used as proxy to measure liquidity position of the company by various researchers. SHARE PRICE BEHAVIOUR Researchers have proposed negative relationship of lagged share prices with current year dividends and positive relationship of current year share prices with dividend distributed during current year. This relationship suggests dividend policy decisions have an impact on shareholders wealth which is mirrored by share prices of a company. CAPITAL EXPENDITURE The extent to which the company decides to finance capital expenditure from internal resources, both dividend and capital expenditure decisions would compete with each other, therefore, capital expenditure is negatively related to its dividend payments. RETAINED EARNINGS A firm that plans to finance future investment opportunities from retained earnings would distribute lesser profits as dividends. Thus, retained earnings of the current year are negatively associated with dividend paid. BETA It measures the systematic risk (systemic risk) of the company. Higher the market risk lower will be the dividend payments. 196

5 PE RATIO There is a debate in corporate finance literature that out of PE ratio and Dividend Payout ratio which is the cause and which is the effect. However in the present study a positive relationship between PE ratio and dividend payout has been assumed. PB or MTBV RATIO: Theoretically higher the growth opportunities available to a firm lower will be the dividend payout. A common proxy used for investment opportunities is MTBV and expected to be negatively correlated with dividend payment. PROMOTER HOLDING It has been recently observed that promoters of the company play a dominant role in dividend decisions of the company. In India promoter companies have earned sizeable income from dividends since dividends are tax free in hands of recipients i.e. promoters of companies in India and dividend forms substantial part of their earnings. SIZE OF THE FIRM Studies have used natural log of total assets and market capitalisation as a surrogate for size of the firm. In particular, larger firms have easier access to external capital markets and can borrow on better terms. Moreover, larger firms tend to be more diversified and their cash flows are more regular and less volatile. Thus, larger firms should be more willing to pay higher dividends. A positive relationship is expected between Dividend payout ratio and firm size as larger firms face lower issuing cost. VOLATILITY IN EARNINGS Volatility in earnings can be measured by taking standard deviations in the earnings per share. An inverse relationship is expected between volatility of earnings and dividend payout ratio. INTEREST COVERAGE RATIO It measures the debt servicing capacity of the firm. A positive relationship is expected between ICR and dividend payout ratio. 197

6 Undeniably all the variables discussed above are important determinants of dividend policy as is evident from the review of the past empirical studies. As such, they need proper consideration by the management while formulating a dividend policy, which will be appropriate both from the point of view of shareholders and that of the company as a whole. This chapter aims at identifying key determinants of dividend policy in three sectors chosen for study. 6.2 DATA AND VARIABLE CONSTRUCTION KEY VARIABLES THAT AFFECT THE DIVIDEND PAYOUT RATIO OF A FIRM On the basis of literature review, the following key variables have been identified that influence the dividend payout ratio of the firm. Y= Equity dividend (in crores), X 1 =PAT (in Rs crore), X 2 =Lagged dividend (Rs. crore), X 3 =Current ratio of firm i during period t, X 4 =Debt equity ratio of firm i during period t, X 5 = Quick ratio of firm i during period t, X 6 = Annual sales growth of firm i during period t, X 7 = Natural log National Stock Exchange adjusted average closing stock prices of the firm i during period t, X 8 = Cashflows of firm i during period t, X 9 = Retained profits of the firm i during period t, X 10 = Capital expenditure or Gross fixed assets (t-(t-1)),x 11 = Nifty beta of firm i during period t,x 12 =Market capitalisation of firm i during period t,x 13 =Price earning ratio of firm i during period t,x 14 =Price to book value ratio of firm i during period t, X 15 = Promoter holding of firm i during period t, X 16 = Natural Log of Total assets of firm i during period t, X 17 = Interest coverage ratio of firm i during period t, X 18 = RONW of the firm i during period t, X 19 = ROE of firm i during period t, X 20 =Lagged PAT (in Rs crore), X 21 = Standard deviation of earnings per share. Therefore, final analysis was carried by reckoning the following standardised key variables. 198

7 Y= dividend payout ratio X 1 =PAT to assets ratio 36 X 2 =Lagged dividend ratio X 3 =Current ratio of firm i during period t X 4 =Debt equity ratio of firm i during period t X 5 = Quick ratio of firm i during period t X 6 = Annual sales growth of firm i during period t 37 X 7 = Natural log National Stock Exchange adjusted average closing stock prices of the firm i during period t X 8 = Cashflows ratio of firm i during period t 38 X 9 = Retained ratio of the firm i during period t X 10 = Capital expenditure or Gross fixed assets (t-(t-1)) to fixed asset ratio) X 11 = Nifty beta of firm i during period t X 12 =Natural log of Market capitalisation of firm i during period t X 13 =Price earning ratio of firm i during period t X 14 =Price to book value ratio of firm i during period t X 15 = Promoter holding of firm i during period t X 16 = Log of Total assets of firm i during period t X 17 = Interest coverage ratio of firm i during period t X 18 = RONW of the firm i during period t X 19 = ROE of firm i during period t X 20 =Lagged PAT to lagged assets ratio (in Rs crore) 39 X 21 = Standard deviation of earning per share 36 In FMCG and Service sector respectively PAT has been expressed as % of total assets. At the same time to obtain better results total assets was substituted by gross fixed assets in IT sector 37 In case of constituent companies of CNX Service sector the annual sales growth was replaced with growth in revenue as majority of the constituents of this Index are banks where sales growth figure is not available 38 In case IT and FMCG sector cashflows have been expressed as a percentage of Netsales. However, in case of Service sector cashflows ratio has been computed by expressing cashflow as a% of PBIT 39 The results in IT sector are reported by expressing lagged PAT as a% of gross fixed assets. 199

8 6.3 IT SECTOR Kaiser-Meyer-Olkin Measure of Sampling Adequacy Table 6.1. depicts KMO values. It is a measure that judges the sampling adequacy. The value obtained is which ensures the sample size is adequate to apply Factor Analysis. The value of KMO is quite encouraging to apply factor analysis. Table 6.1: KMO Test Results: IT Sector Kaiser-Meyer-Olkin Measure of Sampling Adequacy Bartlett's Test of Sphericity Approx. Chi-Square df 210 Sig EXTRACTION METHOD: Principal Component Analysis method was used extraction of factors. The Table 6.2 shows the factor pattern matrix, which highlights variance exhibited by extracted factors SCREE PLOT Scree Plot 6 5 Eigenvalue Component Number Figure6.1: Scree plot :IT sector 200

9 Figure 6.1 shows the scree plot, which is used to determine the number of factors to retain. An elbow in the scree plot indicates the point at which the inclusion of additional factors does not contribute significantly in explaining the variance in data set. Factors above the elbow of the plot are retained. The Scree plot shown above has an elbow at Factor 8.Therefore a set of 8 Factors were chosen which accounts for about 75% of the variations in the data. Table 6.2: Factor Pattern Matrix: IT Sector The table provides the factor loadings from the Principal Component Analysis (PCA). The factor loadings may be viewed as ordinary correlation between a variable and the factor. Underlying loadings in excess of 0.3 are significant FACTOR LABEL Firm size and Liquidity ratios Profitability and Pecking Longterm solvency Leverage and Shareholders wealth and Valuation ratios Dividend stability order retained earnings earnings variability VARIABLES FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FACTOR 6 FACTOR 7 Pat to gross asset ratio Lagged dividend ratio Current Ratio Debt Equity ratio Quick Ratio Sales Growth Ln shareprices CFO to sales ratio Retention ratio Capex to gross fixed asset ratio Beta Mcap PE ratio PB ratio Promoter holding Ln of total assets Interest coverage ratio RONW ROE Lagged profit to lagged fixed assets ratio Standard deviation of EPS Cumulative Variance Explained

10 6.3.4 FACTOR PATTERN MATRIX: The Variables Cashflow from operations to sales ratio, beta, LnMcap, PB ratio, Promoter holding, lagged profit to lagged fixed asset ratio, ROE are heavily loaded on Factor 1. Factor 1 is labelled as Factor of dividend stability and firm size. Higher and more stable the cash flows, higher would be the capacity of the company to pay dividend to the shareholders. In particular, larger firms should have easier access to external capital markets and can borrow on better terms. Even the conflicts between creditors and shareholders are more severe for smaller firms than larger firms. Moreover, larger firms tend to be more diversified and their cashflows are more regular and less volatile. Thus, larger firms are in a better position to pay higher dividends. If the dividend payment of the current year is governed by dividend payment of the previous year, it implies firm follows a policy of paying consistent and stable dividends. Promoters also being one of the major ownership groups have an influence on dividend payout ratio of a firm. It is also commonly observed that firms with larger operations are able to pay consistent and regular dividends. Therefore, a positive relationship is expected between this factor and dividend payout ratio. Variables like Current ratio and Quick ratio are heavily loaded on Factor 2. This Factor has been labeled as Factor of liquidity ratios. Thus, Factor 2 is projected to have positive relationship with equity dividend. PAT to asset ratio, RONW, Capex to asset ratio and sales growth are the two variables that have higher loadings on Factor 3. This factor has been named as Factor of Profitability and Pecking order hypothesis. Dividend and investment decisions of the firm are closely interlinked and cannot be taken in isolation. Higher the growth opportunity available to the firm lesser is the dividend payout of the firm. Lagged dividend ratio and interest coverage ratio (ICR) have significant positive loadings on factor 4. Therefore Factor 4 can be said to represent Factor of long term solvency. Larger the ICR higher is the capacity of the firm to pay stable and consistent dividends. Debt Equity ratio and retention ratio are heavily loaded on factor 5. Therefore; this factor has been termed as Factor of leverage and retained earnings. Higher the risk, more volatile are the earnings and lower is the dividend payout ratios of a company. 202

11 Higher debt equity ratio increases the fixed financial burden of the company in form of interest payments. This in turn puts pressure on firms capacity to pay steady and regular dividends to shareholders. Two arguments support this relationship. First, in an agency theory framework, debt can play a disciplinary role: by increasing the debt level, the free cashflow will decrease (Grossman and Hart 1982; Jensen 1986; Stulz 1990). Indeed shareholders may expropriate wealth from bondholders who try to tackle this problem through indenture restrictions (Jensen and Meckling 1976). Besides, firms with high debt ratios ought to pay lower dividends as they have already precommited their cash flows to make debt payments. Lower dividends also reduce needs to borrow more capital. Thus Dividend payout (DP) ratio of all equity firms is significantly higher than DP ratio of levered firm. Dividend decision of the firm is also closely interlinked with how much profit it expected to retain (Agrawal &Jayram 1994). Hence the Factor 5 is anticipated to be negatively related to dividend payout ratio. Standard deviations in EPS and Ln shareprices have significant positive loadings on Factor 6. Therefore this factor has been labeled as Factor of Shareholders wealth and earnings variability. A firm whose earnings are stable has a better capacity to pay regular and consistent dividend and this leads to creation of shareholders wealth. PE ratio has significant positive loadings on Factor 7. Therefore, this factor can be represented as Factor of Valuation and Capital Market ratios. These ratios reflect the investor s perception of a company. Equity shareholders use these valuation ratios to make investment decisions. It enables them to take hold or exit decisions. Firms with higher PE or PB ratio usually pay generous dividends. Consequently shareholders place higher valuation on share prices of such firms. This factor is expected to be positively associated with Dividend payout ratio RESULTS OF REGRESSION ON EXTRACTED FACTORS The regression coefficients of Factors 1 and 6 have expected signs. Only two factors i.e. Factor 2 & 5 have regression coefficients, which are statistically significant at 5% level of significance. Both factor 2 and 5 have exactly opposite signs of regression coefficients compared to what was expected based on previous research studies. The value of Adjusted R 2 is The value of tolerance is close to 1 which shows that 203

12 there is no problem of multicollinearity in the data. However, multicollinearity is not present in the factor analysis model as indicated by condition index, which is below the critical value 20, and strong stability of coefficients. The Anova Table 6.4 depicts F values, which are significant at 5% level of significance. (For regression results refer to Table 6.3, 6.4 and 6.5). The time dummy variable is also significant at 5% level. The time dummy variable has been introduced to test the structural break in the data set which was expected to prevail as the IT sector was at the bottom of charts in terms of dividend payments till 2004 and in 2005 scenario changed as there was sudden spurt in the dividend payments by the companies in IT sector. Table 6.3: Regression Results on Extracted Factors: IT Sector R R Square Adjusted R Square Std. Error of the Estimate Durbin-Watson Table 6.4 : ANOVA: IT Sector Model Sum of Squares Df Mean Square F Sig. Regression Residual Total Table 6.5: Regression Coefficients of Extracted Factors: IT Sector FACTOR LABEL Regression coefficients Prob.value Condition Index Tolerance (collinearity diagnosis) Firm size and stability Factor Liquidity ratios Factor * Profitability and Pecking order Factor E Long term solvency Factor Leverage and Retained earnings Factor * Shareholders wealth and earnings Factor 6 variability Valuation ratios Factor D1 (Dummy variable for structural break) * Note: * indicates values significant at 5% level of significance 204

13 6.3.6 CONCLUSION This study examines the determinants of dividend payout ratios of companies listed on CNX IT Index in India. The period undertaken for study i.e covers both recessionary and booming phase of Indian information technology sector. Till 2003, there was recession and from 2003, onwards IT sector witnessed exponential growth. After 2006, linear growth has persisted in IT sector. This sector is now, steadily approaching towards maturity. The Return on equity of this sector is very high compared to other sectors of Indian economy. Before 2003 the profitability of the IT firms was scanty and consequently this sector was at the bottom of the list in terms of dividend payments. The average payout of the IT sector during this period was 21.53%. This can be attributed two factors. Firstly, the industry presented immense growth opportunities for the companies hence the managers were of opinion that they can provide the investors better returns if they plough back the earnings into business. Secondly, most firms in Industry were facing volatile earnings stream, which deterred them from paying more dividends. After 2003, there was a substantial spurt in dividend payout ratios of the IT companies. Infosys Technologies, Wipro Technologies, HCL Technologies were among the highest dividend paying companies. Infosys Technologies paid dividend as high as 2590% in the year The surprising results are that factor 2 and 5 have exactly opposite signs in contrast to what was expected based on literature survey. This implies that firms in IT sector do not use dividends as a medium to signal their prosperity to the shareholders. It also reflects that there is lesser information asymmetry in this sector. The information is becoming more and more symmetrical due to better Corporate Governance practices adopted by IT companies. IT sector is a human intensive sector and do not require huge capital asset base like manufacturing companies for their operations. The major asset of this sector is manpower. The funds required for recruitment and retention of manpower is comparatively less than funds required for purchasing capital assets. So these firms can easily release funds for payment of dividends. Also a negative relationship between liquidity can be attributed to the fact that agency problems are not very relevant so that monitoring mechanism i.e. dividend payout may be less needed. A 205

14 negative regression coefficient of Factor 2 and DP ratio can be attributed to the fact that in IT sector capital gains are preferred to cash dividends. Retained earnings are a vital variable governing the Dividend Payout ratio of IT firms. The results show that Factor of retained earnings and Leverage is positively related to DP ratio. Generally, higher debt equity ratio may negatively influence the dividend payout of company. But in case of IT firms the proportion of debt in the total capital structure of the company is relatively low as they are very low debt or zero debt companies eg. Infosys ( a zero debt company). Therefore, bondholders do not consider dividend payment a way to expropriate their value. This relationship also highlights lesser conflicts between two groups of stakeholders i.e. shareholders and bondholders in the Indian IT sector. Stockholders may expropriate wealth from bondholders by paying themselves dividends. But the results show that Bondholders may also try to contain this problem through restrictions on dividend payments in bond indenture. Thus, the positive relation depicts that debt holders do not reduce the cash available for the dividend by imposing debt covenants and related restrictions. This positive relationship between Dividend payout ratio and Debt equity ratio is in alignment with the findings of Easterbrook (1984). According to him firms with high leverage are those whose value shifting is potentially costly. Such firms are expected to pay large dividends. The Factor of profitability and pecking order, long-term solvency, Shareholders wealth and earnings variability, have not emerged as an imperative factors affecting the dividend payout ratios of IT firms The existing variables explain just 37% of Indian Information Technology dividend behavior; future research can be focused on discovering variables that can offer better explanation of dividend payments. 206

15 6.4 FMCG SECTOR Kaiser-Meyer-Olkin Measure of Sampling Adequacy KMO was calculated as a first step. It is a measure that judges the sampling adequacy. The value obtained is.691 which ensures the sample size is sufficient to apply Factor Analysis. Table 6.6 below shows the test values. Table 6.6: KMO test values: FMCG SECTOR Kaiser-Meyer-Olkin Measure of Sampling Adequacy Bartlett's Test of Sphericity Approx. Chi-Square Df 210 Sig EXTRACTION METHOD: Principal Component Analysis method was used for extraction of factors. The Table 6.7 shows the factor pattern matrix, which highlights variance exhibited by extracted factors SCREE PLOT Scree Plot 6 5 Eigenvalue Component Number Figure 6.2: The Scree plot: FMCG sector The Scree plot shows the factor eigen values in descending order.the eigen values of a factor represents the variance explained by each factor. An elbow in the Scree plot occurs at Factor 6, which indicates the point at which the inclusion of additional 207

16 factors does not contribute significantly in explaining the variance of the data set. The results of the analysis are presented in the form of factor pattern matrix. Factors above the elbow of the plot are retained. A set of 6 Factors that were chosen accounts for about 76% of the variations in the data FACTOR PATTERN MATRIX: Table 6.7: Factor Pattern Matrix: FMCG Sector The table provides the factor loadings from the Principal Component Analysis (PCA). The factor loadings may be viewed as ordinary correlation between a variable and the factor. Underlying loadings in excess of 0.3 are significant FACTOR LABEL VARIABLES Dividend signaling and smoothing Cash flow quality and firm size Systematic and financial risk Pecking Order hypothesis Liquidity ratios and ownership dispersion Longterm solvency and shareholders wealth FACTOR 1 FACTOR 2 FACTOR 3 FACTOR 4 FACTOR 5 FACTOR 6 Pat to total asset ratio Lagged dividend ratio Current Ratio Debt Equity ratio Quick Ratio Sales Growth Ln shareprices CFO to sales ratio Retention ratio Capex to gross fixed asset ratio Beta LnMcap PE ratio PB ratio Promoter holding Ln of total assets Interest coverage ratio RONW ROE Lagged profit to lagged total assets ratio Standard deviation of EPS Cumulative Variance Explained

17 An analysis of the factor pattern matrix potrays that the variables like PAT to asset ratio, lagged DP ratio, lagged PAT to asset ratio, ROE and RONW have significant positive loadings on Factor 1. Factor 1 is labelled as Factor of dividend signaling and smoothing. Dividend carries a signal of firm s prosperity. Shareholders tend to applaud dividend increases and frown dividend cuts. Several research studies have convincingly proven that dividend has information content. Higher the profitability, higher is the dividend payout of the firm. A firm strives to maintain stability and consistency in dividend payout if Dividend Payout ratio of the current year is governed dividend paid during previous and profit of the current year Therefore, a positive relationship is expected between this factor and dividend payout ratio. Variables like Cashflow from operations to sales ratio, LnMcap and Lntotal assets are heavily loaded on Factor 2. This Factor has been labeled as Factor of cash flow quality and firm size.a negative loading of cashflow to sales ratio on this factor depicts that firms with larger size suffer from liquidity problems and shortage of cash. Higher and more stable the cash flows, higher would be the capacity of the company to pay dividend to the shareholders. Generally larger the firm size greater is the capacity of the firm to pay dividend.thus, Factor 2 is projected to have positive relationship with equity dividend. Variables like Debt equity ratio, Beta and Standard deviation in earnings per share are heavily loaded on factor 3. This factor has been named as Factor of financial risk and systematic risk. Higher the risk, more volatile are the earnings and lower is the dividend payout of a company. Higher the risk, more volatile are the earnings and lower is the dividend payout of a company. Higher debt equity ratio increases the fixed financial burden of the company in form of interest payment. This in turn puts pressure on firms capacity to pay steady and regular dividends to shareholders. Thus factor 3 is projected to show a negative relationship with DP ratio. Variables like sales growth, Capex to asset ratio, retention ratio and lagged dividend payout ratio have significant positive loadings on factor 4. Therefore this Factor 4 can be said to represent Factor of Pecking order Hypothesis. Higher the growth opportunities available to a firm lower will be dividend payout and lesser would be the firms capacity to pay consistent dividends. Therefore, the lagged DP ratio has a 209

18 negative loading on factor 4. Thus, Factor 4 is expected to have negative relationship with Dividend Payout ratio. Current ratio, Quick ratio display significant positive loadings on Factor 5, indicating that liquidity and promoter holding positively impact dividend payout. It can also be interpreted in the sense that firms with high promoter holding are better managed and enjoy higher liquidity. This factor is labeled as Factor of Liquidity ratios and Ownership dispersion. Lnshareprices, interest coverage ratio have higher loadings on Factor 6. Therefore this factor can be termed as Factor of longterm solvency and shareholders wealth. Higher is the ICR higher is the firm s capacity to pay dividends. At the same time higher debt equity ratio induces firm to pay regular dividends to the shareholders to mitigate agency conflicts. Also shareholders wealth as proxied by Lnshareprices are positively loaded on this factor depicting that a firm, which is highly solvent in long term, enjoys higher share prices. Hence the Factor 6 is anticipated to be positively related to dividend payout ratio RESULTS OF REGRESSION ON EXTRACTED FACTORS The regression results have been reported in Table 6.8, 6.9 and 6.10 respectively. The regression results indicate that Factors 1,2,4 and 6 have expected signs of regression coefficients. Out of these factors only one factor i.e. Factor 6 has regression coefficient, which is statistically insignificant at 5% level of significance. Factor 5 has exactly opposite signs of regression coefficient compared to what could be expected based on previous research studies. The value of Adjusted R 2 is 32.8% The value of tolerance is close to 1 which shows that there is no problem of multicollinearity in the data. This is also confirmed by the fact that the value of condition Index is below the critical value of 20. The DW statistics is The Anova table 6.9 depicts F values which are significant at 5% level of significance. The dummy variable has also been introduced to take care of structural break that is anticipated from and The FMCG sector revived from prolonged slump in the beginning of 2005 and a change in 210

19 the taxation structure with the introduction of VAT had several implications for the sector 40. Table 6.8: Regression Results on Extracted Factors: FMCG Sector Adjusted R Std. Error of the R R Square Square Estimate Durbin-Watson Table 6.9: ANOVA: FMCG Sector Model Regression Sum of Df Mean Square F Sig. Squares Residual Total Table 6.10: Regression Coefficients of Extracted Factors: FMCG Sector FACTOR LABEL Regression coefficients Prob.value Condition Index Tolerance (collinearity diagnosis) Dividend signaling and Factor 1 smoothing * Cash flow quality and firm size Factor * Systematic risk and financial risk Factor * Pecking order hypothesis Factor * Liquidity ratios and ownership dispersion Factor of long-term solvency and shareholders wealth D1 ((Time dummy for structural break) Factor 5 Factor ** Note: * and ** indicates values significant at 5% and 10% level of significance respectively. 40 For details refer to the Industry overview chapter 211

20 CONCLUSION The period undertaken for study ( ) covers a complete business cycle of Indian FMCG sector. This sector recovered from its prolonged slump in 2005 FMCG companies have been known to be generous dividend distributors to its shareholders. These stocks were known as dividend yield stocks till 2004.The companies maintained consistent dividend payouts to some extent even when the profits were not on surge. After 2004, FMCG stocks were purely viewed as dividend growth stocks since the companies deployed their resources for sustaining larger product baskets. It exposed the companies to greater competition. The FMCG companies adopted a CAPEX mode and started ploughing profits for future expansion plans. But the high profitability enabled the firms to continue pay encouraging dividends even after retaining a part of the profits. HLL, Godrej, ITC are among the top dividend payers. FMCG companies business is easy to understand, as it is relatively simple and stable in character. A complex business causes difficulty in predicting future cash flows. The products have a quick turnover, and relatively low input cost. The consumers think less while purchasing FMCG products as they are meant for daily requirements. FMCG companies typically require very little incremental Capex. They do not have to invest huge new capital in assets to grow earnings. That is what makes these companies so discreet in terms of capital efficiency. Further, they also have strong cash flows and a low debt to equity ratio. Most of the FMCG and food companies are debt free, thus are not affected by interest rate cycles. Companies with stable, simple and have sustainable competitive advantages over peers are likely to generate materially higher cash flows with the passage of time. The FMCG companies usually fulfill these criterions. Their strong brands and multiple product innovations help them sustain their revenue stream over long periods. Also, the consumers buy the same product several times in a year. FMCG companies are known to be generous payers of dividend due to their strong cash flow and minimal Capex requirement. Indian FMCG companies like their global peers have developed some strong brands, sustained stable growth, high dividend payout and high return on net worth (RONW). It can be stated from above analysis that Profitability is a primary determinant factor for dividend distribution. FMCG companies score high on dividend stability and consistency, as Lagged dividend and PAT are important factors governing dividend 212

21 distribution. These findings are in alignment with the findings of Aharony and Wary (1980), Asquith and Mullins(1983), Petit(1972), John and Williams(1980), Bhattacharya(1979), Miller and Rock(1985). The quality of cash flows, which is a measure of liquidity of the firm and firm size are found be to be a note worthy determinant of the dividend payout. According to previous research studies larger firms face lower issuing cost and external debt financing making it easier to raise funds. Thus, they can go for generous dividend payouts. These findings are consistent with the findings of Smith and Watts (1992)[118]. The opportunities for future growth and expansion are found to be negatively related to dividend payout ratio. Larger is the growth and investment opportunities available to the firm, lesser is the incentive to pay dividends as the firms prefer to retain larger proportion of profits. According to Pecking order hypothesis, firms should prefer to finance investment by retentions rather than debt. The regression results also disclose negative and significant relationship with Retained earnings and Capital Expenditure during the current year. This result is in alignment with the existing literature, which suggests that results are logically, and theoretically correct. A company, which prefers retention of profits for financing the capital expenditure from internal resources, distributes fewer dividends compared to a firm, which finances the investment expenditure from external sources. Thus, the extent to which the company decides to finance CAPEX from retained earnings; both dividend and CAPEX in a company would be negatively related to dividend payments. In other words, dividend decisions are not independent of uses of corporate funds and changes in fixed assets i.e. capital expenditure is an important determinant of dividend payments in FMCG sector. The results depict a negative relationship between liquidity and Dividend Payout ratio and promoter holding. This relationship cannot be validated as regression coefficient of Factor Liquidity ratios and ownership dispersion is found to be insignificant at 5% level. The systematic risk, earnings variability and financial risk obstruct the stable dividend payout but the results report that in case of FMCG sector in India the Dividend Payout ratio is increasing even if the firm faces higher risk. 213

22 Dividend Payout ratio is found to be positively related to long term solvency of the firm. But this relationship is significant at 10% level. The firms in FMCG sector operate with very low level of debt. At the same time these firms are highly liquid firms, therefore increase in debt proportion in capital structure do not put pressure on firms capacity to pay dividend and consequently a positive relation can exist between Debt Equity ratio and Dividend Payout ratio through the results. Dividend are less important to investors in high growth firms who seek out these firms in the expectation of receiving little dividend income as these firms need outside financing regularly and therefore are subject to the discipline of frequent capital market scrutiny. This also indicates lesser degree of conflicts between bondholders and shareholders. The bondholders do not reduce the cash available for dividend distribution by imposing indenture restrictions SERVICE SECTOR Kaiser-Meyer-Olkin Measure of Sampling Adequacy : The first step was to calculate KMO. It is measure that judges the sampling adequacy the value obtained is which ensures the sample size is ample to apply Factor Analysis. Table 6.11: KMO Test Values: Service Sector Kaiser-Meyer-Olkin Measure of Sampling Adequacy Bartlett's Test of Sphericity Approx. Chi-Square df 210 Sig..000 Bartlett test of spherecity is the statistical test for overall significance of all correlations with in a correlation matrix. It judges the appropriateness of factor analysis Extraction Method: Extraction method used is Principal Component Analysis. The Table 6.9 highlights the variance exhibited by extracted factors. It shows that the first factor accounts for highest amount of variance, the second factor accounts for second highest and so on. The principal components analysis using varimax rotation method of correlation matrix of the 22 variables have led to the extraction of seven 214

23 broad components of dividend policy of the corporate India which are explained in the forthcoming paragraphs. These factors accounted for 20%, 12%, 10%, 10%, 7%, 7% and 5% of the total variance explained respectively SCREE PLOT The scree plot is used to determine the number of factors to retain. Factors above the elbow of the plot are retained. The procedure involves certain amount of subjectivity if no clear elbow appears in the curve. The Scree plot shown below shows a clear elbow at Factor 7. These seven factors cumulatively accounts for about 72% of the variations in the data. Consequently, these seven Factors are considered for the analysis. Scree Plot 5 Eigenvalue Component Number Figure6.3: The Scree plot: Service sector 215

24 Table 6.12: Factor Pattern Matrix: Service Sector The table provides the factor loadings from the Principal Component Analysis (PCA). The factor loadings may be viewed as ordinary correlation between a variable and the factor. Underlying loadings in excess of 0.3 are significant FACTOR LABEL Dividend signaling and profitability Liquidity ratios and systematic risk. Agency conflicts and ownership dispersion Firm size and dividend stability Cash flow quality and valuation ratios Pecking Order hypothesis Long-term solvency and earnings variability VARIABLES FACTOR 1 FACTOR 2 FACTOR 3 FACTOR FACTOR 5 FACTOR 6 FACTOR 7 4 Pat to total assets ratio Lagged profit to lagged total assets ratio Lagged dividend ratio Current Ratio Debt Equity ratio Quick Ratio Growth in total Inome Ln shareprices CFO to EBIT ratio Retention ratio Capex to gross fixed asset ratio Beta LnMcap PE ratio PB ratio Promoter holding Ln of total assets Interest coverage ratio RONW ROE Standard deviation of EPS Cumulative Variance Explained FACTOR PATTERN MATRIX: The results of the analysis are presented in the form of factor pattern matrix. The Variables PAT to asset ratio, PB ratio, Beta, ROE and RONW Lagged PAT to asset ratio have significant positive loadings on Factor 1. Factor 1 is labelled as Factor of dividend signaling and profitability. As it is known that dividend carries 216

25 a signal of firms prosperity. Shareholders tend to welcome dividend increases and scowl dividend cuts. Thus, dividend has information content. Higher the profitability, higher is the dividend payout of the firm. Dividend is a medium to signal profitability of the company to the sharholders. Dividend stability exists when dividend paid during current year is governed by previous year profits and dividend paid during the year t-1 (Lintner 1956). Therefore, a positive relationship is expected between this factor and dividend payout ratio. Variables like Current ratio and Quick ratio are heavily loaded on Factor 2. These ratios are proxies for firms liquidity. Also beta, a proxy for systematic risk and standard deviation in earnings have significant positive loadings on the same factor indicating that firms with higher systematic risk tend to maintain higher liquidity.the variability in earnings leads to fluctuations in the dividend paid. Firms having higher systematic risk are expected to show greater variability in the earnings. If the issuing costs are significant, then firms are likely to finance investments through retention of earnings rather than from external sources. To the extent that dividends compete with investments for internally generated funds, such costs are likely to affect dividend policy. Hence under the residual theory of dividends a negative relationship is expected between dividend payout and external financing costs. Rozeff (1982) has used equity beta to proxy for the cost of external financing. Therefore a negative relationship is expected between beta and Dividend Payout ratio. This Factor has been labeled as Factor of liquidity ratios and systematic risk. Previous research studies have shown both positive and negative signs of the variables loaded on this factor with dividend payout ratio. Debt Equity ratio, promoters holding are heavily loaded on factor 3. Debt Equity ratio is positively loaded while the promoter holding is negatively loaded. This implies firms where promoter holding is high tend to have lower debt equity ratio. Therefore, this factor has been termed as Factor of agency conflict and ownership dispersion. Higher the risk, more volatile is the earnings and lower is the dividend payout of a company. Highly Levered firms can write strict dividend policy constraints. Heavy debt service obligations limit these firms ability to over invest and frequent refinancing provides capital market discipline. The same time higher the money raised through debt lead to conflicts between bondholders and equity shareholders. 217

26 Therefore, the firm is required to pay higher dividends to shareholders to mitigate the agency conflicts. The financial risk proposition as proxied by Debt Equity ratio might suggest that firms with greater financial risk ought to pay higher dividends to investors in order to compensate for risk. Therefore this factor is estimated to have positive relationship with dividend payout ratio. Lnshareprices, Lagged dividend payout ratio retention ratio, LnMCap and Lntotal assets have significant positive loadings on factor 4. It has been seen larger the firm size higher is the dividend payout of a company. It may be expected that smaller firms grow faster through retention and so there would be a negative relationship between retention ratio and firm size, and hence a positive relationship between the dividend payout ratio and firm size. Larger firms are in a better position to maintain stable dividends and create value for shareholders as reflected by the shareprices of a company.therefore, Factor 4 can be said to represent Factor of firm size and stability. Factor 3 is projected to have positive relationship with equity dividend. It can also have negative sign with the Dividend Payout ratio as it is observed larger firms face larger cost of external financing thus prefer to retain funds and pay lower dividends. Variables like, cash flow to EBIT ratio and PE ratio have higher loadings on Factor 5.Therfore this factor can be termed as Factor of cash flow quality and valuation ratios. If the dividend payout of the current year is governed by dividend payout of previous year, it implies firm follows a policy of paying consistent and stable dividend. Dividend decision of the firm is also closely interlinked with how much cash flows are generated from operating activities. Hence the Factor 5 is anticipated to be positively related to dividend payout ratio. Capex to asset ratio and growth in total income are the two variables that have higher loadings on Factor 6. This factor has been named as Factor of Pecking Order Hypothesis. Dividend and investment decisions of the firm are closely interlinked and cannot be taken in isolation. Higher the growth opportunity available to the firm lesser is the dividend payout of the firm. According to pecking order hypothesis, firms should prefer to finance investment by retentions rather than debt. A higher retention ratio implies a lower dividend payout ratio, so lower payout ratio should be associated 218