The Determinants of Corporate Dividend Policy
One of the central issues of corporate finance has been the dividend decision of a firm, which has always been studied in relation to a firm’s financing and investment decisions. The association amongst these two decisions has posed various questions. How much should a firm pay as dividend? How does a dividend payout policy influence the valuation of a firm? Does a firm’s decision to distribute cash correspond to its financing and investing decisions?
What is the outcome of changes in the dividend policy assuming steady financing and investment decisions of a firm? Research has attempted to provide answers to these questions and many more but mystery still shrouds the dividend decision. Lintner (1956) argues that firms of developed markets target their dividend payout ratio with the help of current earnings and past dividends. In order to reach such target, various adjustments are made in the dividend policy of a firm and therefore firms should have stable dividend policies.
The Determinants of Corporate Dividend Policy Essay Example
Miller and Modigliani (1961) on the other hand feel that dividend policy is irrelevant in measuring the current worth of shares considering the irrational assumptions of market perfections, zero transaction costs, perfect certainty and indifferent behaviour of investors. However, Miller and Scholes (1982) argue that in the real world, dividend decision is inspired more by high taxes on dividends than capital gains and market imperfections. Alli, Khan and Ramirez (1993) observe that a change in the payout policy provides information about future earnings and a further change in the value of share price.
This indeed shows a strong signaling effect of the dividend decision of a firm. It is evident that over the years, no Dr. Amitabh Gupta, Associate Professor, Department of Financial Studies, University of Delhi Ms. Charu Banga, Research Scholar, Department of Financial Studies, University of Delhi Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 64 single viewpoint has emerged which explains the dividend policy of a firm. In India too, modest research has been carried out on various aspects of the dividend decision.
The present study re-examines the impact of various factors on the dividend decision of Indian companies taking a large and latest data set. Our study contributes to the existing literature by examining as many as fifteen financial variables. No prior Indian study has examined so many variables in the context of dividend decision using both factor analysis and regression. The remaining paper is organized as follows: Section 2 presents the existing literature.
Section 3 discusses the hypothesis being tested in the study and also describes the data and the research methodology used in the study. Section 4 discusses the results of the study and section 5 presents the conclusion. Literature Review A large number of financial and non-financial determinants of corporate dividend policy have been discussed in the work of Lintner (1956). This seminal work developed a basic model stating that most of the companies follow dividend adjustment process by applying target dividend payout ratio.
Rozeff (1982) investigates the impact of two kinds of costs – transaction costs and agency costs relative to external financing on the dividend decision of a firm. He argues that a balance between transaction costs and agency costs would lead to an optimum dividend policy. Alli, Khan and Ramirez (1993) find that dividends do not convey information regarding a firm’s future cash flows. They report that at beta, firm’s capital expenditure and financial slack are inversely related to the dividend payout.
The dividend policy behaviour is also examined by Han, Lee and Suk (1999) by considering institutional ownership under agency cost hypothesis and tax–based hypothesis. They find that tax-based hypothesis is more relevant in the case of institutional investors as they prefer a greater dividend payout. Pandey (2001) looks at the corporate dividend payout behaviour of companies listed on the Kuala Lumpur stock exchange during 1993-2000. He categorizes the sample into six industries for examining the variation in the payout ratio. He also establishes a relationship between current earnings and past dividend rate.
He finds that the Malaysian companies (by following Lintner’s model) exhibit unstable dividend behaviour with high adjustments in dividend payments in order to meet the target payout ratio. Myers (2004) finds strong support for earnings, profit margin, institutional ownership and debt-equity ratio on the dividend decision. Eriotis (2005) finds that Greek firms have a long-run constant dividend payout policy. He adjusts the firms’ distributed earnings and size in the Lintner model and reports that an increase in the earnings does not change the dividend distribution pattern of firms.
Kania and Bacon (2005) find that variables such as sales growth, expansion and insider ownership have a negative impact on dividend decision but institutional ownership has an inverse relation with dividend payout, which is contrary to the existing literature. Denis and Osobov (2008) find that the tendency for paying dividends declined for countries such as United States, Canada, United Kingdom, Germany, France and Japan during 1994-2002. They also report that the international evidence does not support the investors’ preference for dividend, the signaling and the clientele interpretations as prominent variables.
Rather, they go along with the distribution of free cash flow as the chief element of the dividend decision. Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 65 In the Indian context, we found few studies that have analyzed the factors that affect the dividend decision of a firm. For example, Kevin (1992) analyzes the dividend payment behaviour of 650 Indian companies during September 1983 to August 1984 and finds that profitability and earnings of the firms are the two foremost factors determining dividends.
He concludes that Indian firms strive for achieving a stable dividend rate. However, keeping in view that the time period of his study was only one year; his results cannot be taken as conclusive. Mahapatra and Sahu (1993) find that cash flows, current earnings and past dividends are prominent factors that have an impact on the dividend decision. Their results are in contrast to Lintner’s model. Bhat and Pandey (1994) find that current year’s earnings, pattern of past dividends, expected future earnings, changes in equity base of the firm have an impact on the dividend decision.
Taking a different line of research, Narasimhan and Asha (1997) look at the changes in dividend tax regime proposed in the Indian Union Budget of 1997-98 and analyze the impact of dividend tax on a firm’s dividend decision. They conclude that the burden of tax payment fell in the hands of companies rather than their shareholders. Mohanty (1999) studied more than 200 Indian companies for a period of fifteen years to understand the relationship between bonus-issuing and dividend-paying behaviour of companies.
He found that in the Indian context, it is the dividend rate that is an important determinant of dividend policy in comparison to the dividend payout ratio. Reddy (2002) analyzes the trends and determinants of dividend of all Indian companies listed on two major Indian stock exchanges–The Bombay Stock Exchange (BSE) and The National Stock Exchange (NSE) during 1990-2001. He investigates three factors viz. , number of firms paying dividend, average dividend per share and the average payout. His results indicate that only few companies maintain the dividend payout rate and that firms forming a part of small indices pay higher dividend compared to firms forming a part of broad market indices. Deviations in the tax regime are also examined using the trade-off theory and it is found that this theory does not apply to the Indian corporate sector. He concludes that the omission of dividends have information content i. e. such companies expect lower earnings in the future whereas the same does not hold true in case of dividend initiations. Anand (2004) analyzes the results of Anand (2002) survey of 81 CFOs to find out the determinants of dividend policy of Indian companies.
He finds that Indian companies use dividend policy as a signaling mechanism to convey information about their present and future prospects, therefore, affecting their market value. He also reports that while designing a dividend policy, companies take into consideration the investors’ preference for dividends and the clientele effect. The relationship between corporate governance and dividend payout behaviour of the Indian firms is examined by Kumar (2006) by taking into consideration their financial structure, investment opportunities, dividend history, earnings trend and ownership structure during 1994–2000.
He finds a positive association of dividends with earnings and dividend trends but does not find any association between foreign ownership and growth in dividend payout. Recently, Bhayani (2008) has examined the influence of earnings and lagged dividend on dividend policy of companies listed on the BSE. He found that the current year’s earnings is the foremost factor affecting the dividend behaviour of a firm and concludes that Indian Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 66 companies follow a stable cash dividend policy.
Kanwal and Kapoor (2008) examine the dividend policies of companies in the information technology sector in India. They explore various factors such as profitability, cash flows, corporate tax, sales growth and growth opportunities that have an impact over the dividend policies of such companies. They report that only cash flows indicating liquidity and beta indicating risk are the foremost determinants. Thus over the years different strands of research have emerged in the area of dividend policy both in India and abroad. Data and Research Methodology Data and Sample Selection
The necessary data have been sourced from the Prowess database of Centre for Monitoring Indian Economy (CMIE)1 The sample companies are drawn from the broad based BSE 500 Index2. The period of the study is seven years from January 1, 2001 to December 31 2007. We only included those companies in the sample that had continuously paid dividend during the study period and have excluded financial institutions/finance companies and government owned companies. Only final cash dividends paid by companies have been considered as usually Indian companies pay only one dividend during a year.
We have also ignored stock dividends and stock repurchases by companies and have examined only cash dividend. This process gave us a final sample of 150 companies from 16 industries. Description of Variables Over the years researchers have employed numerous financial variables that have a possible impact on the dividend policy (A list of such variables is provided in Annexure 1). Out of such variables, the present study considers fifteen variables to examine their effect on the dividend decision. The justification for choosing such variables is as follows. Liquidity is an important determinant of dividend decision.
Liquidity and dividend payment behaviour of a company have a direct relationship (Benito and Young, 2001). If a company has adequate cash flows, it would like to distribute cash dividend in order to keep its shareholders happy. Moreover, firms have to make their dividend payments in cash therefore they have to be liquid enough to distribute dividends and also to remain solvent. Current ratio (CR) and cash from operations (CFO) are the indicators of the liquidity position of a firm. Thus, CR and CFO become the first and second variable respectively. Another significant factor is leverage.
A firm with high leverage means large fixed payments for external financing, which indeed is a substitute for the dividend payments. High leverage increases the transaction costs and the risk of the firm (Rozeff, 1982). Contrarily, higher the earning retention rate, lower the chances for external 1 Prowess is a database containing information about large and medium Indian companies and is commonly used by both the researchers and the industry alike in India 2 Bombay Stock Exchange (BSE) is one of the largest stock exchanges of Asia. Its broad based index – BSE 500 constitutes major blue chip companies in India
Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 67 borrowing and vice-versa. Hence we take debt-equity ratio (DER) and ratio of retained earnings to equity (REE) as proxies for financial leverage exhibiting a negative relationship with the dividend decision. Hence, DER and REE become the third and fourth variable respectively. Further, the relationship between ownership structure and dividend payment behaviour of a firm is also valuable. The control of the firm may lie with the directors or the promoters (insider owners), institutions (institutional owners) or with foreign investors.
The insiders would like to avoid excess payment of dividend whereas institutional owners are usually more dividend demanding (Han, Lee and Suk, 1999). Thus, promoters’ shareholding (PS), institutional shareholding (IS) and foreign institutional investors shareholding (FIIs) represent the fifth, sixth and seventh variables respectively. Profitability has always been considered as the foremost determinant for dividend payment as more profit means more dividends. It becomes essential to consider variables for short-term and long-term profitability of a firm (Myers, 2004). We take net profit ratio (NPR), return on investment (ROI) and ratio of profit before interest & tax to total assets (PTA) as their proxies and therefore, they become the eighth, ninth and tenth variables respectively. In addition, growth opportunities play an important role. Higher the operational growth and growth in profits of a firm, higher shall be the dividend payments by the firm (Kania and Bacon, 2005). The growth factor is represented by annual sales growth (ASG), return on net worth (RONW) and earnings per share (EPS) growth. The growth rates of ASG, RONW and EPS are taken as the eleventh, twelfth and thirteenth variables respectively.
Market capitalization (MC) corresponds to the size of the firm and is the fourteenth variable. Tax effect is another significant determinant as the rates of taxation influence the need for dividend by the investors. High tax paying investors would prefer to postpone receiving dividends and thus, would like to retain their earnings with the firm in order to avoid heavy taxes whereas investors in lower tax brackets would prefer higher dividends (tax clientele effect)3. For this purpose we take ratio of corporate tax to profit after tax (T) as a proxy and the fifteenth variable. A detailed definition of these variables is given in Annexure 2.
Research Methodology The present study re-examines various factors that have a bearing on the dividend decision of a firm by using a two-step multivariate procedure. We have identified fifteen variables from the extant literature that are considered while framing a dividend policy. In the first step, we perform factor analysis on the data to extract prominent factors from these fifteen variables. In the second step we perform multiple regression on the factors extracted. Results Factor Analysis The technique of factor analysis indicates those factors that explain the correlation among the observed variables.
We use principal component analysis (PCA) as the factor extraction method to identify distinct clusters of observed variables (Dillon and Goldstein, 1984; Tryfos, 3 Tax Clientele Effect: See Miller and Modigliani, 1961. Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 68 1998). The broad factors are further subjected to equamax orthogonal rotation (Alli, Khan and Ramirez, 1993). Table 1 shows the results of Kaiser-Meyer-Olkin (KMO) and Bartlett’s Test. The KMO measure of sampling adequacy tells us whether we can reduce the variables into broad factors or not.
Value of less than 0. 50 indicates that factor analysis would not produce distinct and reliable factors whereas any value close to one would generally indicate that this technique of analysis would be useful with the data. Our results gave a value of 0. 554 indicating that the pattern of correlation amongst the variables is relatively compact and hence, Factor Analysis yields distinct and reliable broad factors (Meyers, Gamst and Guarino, 2006). The Bartlett’s test of Sphericity investigates whether the original correlation matrix is an identity matrix or not.
Our results show that Bartlett’s test has a chi-square value of 1. 500E3 which is significant for p < 0. 01 confirming that factor analysis is appropriate. Table 1: KMO and Bartlett’s Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy. Bartlett’s Test of Sphericity .554 Approx. Chi-Square 1. 500E3 Df 105 Sig. .000 Table 2 gives the Rotated Factor Matrix using Equamax Orthogonal Transformation i. e. a matrix of factor loadings for each variable upon each factor. The factor loadings of less than 0. 30 have been suppressed and are not displayed. The institutional shareholding (0. 853) and foreign institutional investors’ shareholding (0. 735) are positively correlated whereas promoters’ shareholding (-0. 824) is negatively correlated to the first factor i. e. ownership structure. The dividend payout tends to bring a decline in the stock value, thus, a conflict of interest for the insiders. A company with high insider ownership proposes for a low cash dividend payout. Whereas, institutional owners are keen to influence high payouts in order to enhance control over the management for monitoring their external financing matters (Kumar 2006, Myers, 2004, Han, Lee and Suk, 1999).
Our results strongly support the findings in the literature. However, one point worth noting is that the individual shareholdings of promoters, institutions and foreign institutional investors in relation to the total shareholdings of a firm have not been taken into consideration. This can be an area for future research. Factor 2 has high negative loadings for debt-equity ratio (-0. 756), taxation (-0. 704) and earning per share (-. 384) and positive loadings for return on investment (0. 759) and return on net worth (0. 577). We term this factor as leverage.
This suggests that firms would like to pay high dividends if they are utilizing their retained earnings (least risk attached) as compared to external financing (equity and debt). In other words, high interest payments (fixed charge) will result in lower dividend payment (Alli, Khan and Ramirez, 1993 and Rozeff, 1982). Therefore, results indicate that there exists an inverse relationship between dividend rate and leverage. Decision, Vol. 37, No. 2, August, 2010 The Determinants of Corporate Dividend Policy 69 The third broad factor is expressed as Profitability. It includes net profit ratio (0.932), ratio of profit to total assets (0. 932) and current ratio (0. 546). This factor indicates that greater the profit of a firm, higher will be the dividend payout. Therefore, profitability is positively related to dividend decision (Denis and Osobov 2008, Myers 2004). Our results confirm the same. The fourth factor incorporates high positive loadings for cash from operations (0. 809) and market capitalization (0. 789). We coin this factor as liquidity. A firm with high external financing would require availability of cash flows i. e. strong liquidity position to meet its financial obligations.
Therefore, in order to increase liquidity, the firm shall lower its dividend payout. On the other hand, the larger the size of the firm, the greater the availability of free cash flows and the greater will be the dividend payout. A firm with large number of shareholders is expected to pay higher dividends in order to keep their shareholders happy. It has been found that high retained earnings to equity ratio (indicating propensity to pay dividend) would ensure availability of free cash flows or residual cash flows within the firm (Benito and Young, 2001). One would, therefore, expect a direct relation between liquidity and dividend payout.
The fifth factor is termed as Growth. It includes annual sales growth (0. 816), return on net worth (0. 530) and ratio of retained earnings to equity (0. 485) implying that growth in sales and profit is an important determinant for the payment of dividends. Our results support the findings of Myers (2004) who suggests that firms with high growth rate distribute high dividends in order to keep their shareholders happy. Table 2: Rotated Component Matrix Components IS PS FIIs ROI DER T RONW EPS PTA NPR CR CFO MC ASG REE Ownership Structure .843 -. 824 .735 Leverage Profitability
Liquidity Growth .385 .759 -. 756 -. 704 .577 -. 384 Decision, Vol. 37, No. 2, August, 2010 .412 .415 .932 .932 .546 .530 -. 431 .809 .789 .816 .485 The Determinants of Corporate Dividend Policy 70 Regression Results In the second step, multiple regression is carried out to examine the impact of the five independent variables on the dividend rate. The dividend rate is a dependent variable constituting the dividend decision and the five factors extracted from factor analysis viz. leverage, liquidity, profitability, growth, and ownership structure are taken as the independent variables.
Since the factors used in the regression model are derived through the orthogonal transformations, they are free from multi-collinearity problems (Ali, Khan and Ramirez, 1993). Further tests for normality, heteroscedasticity and autocorrelation show that data is normally distributed and there are no related problems. Table 3 gives the results of the regression model. The R-square is 0. 244 i. e. around 25 per cent of the variability in dividend rate is explained by the independent variables tested. The FStatistic of 9. 320 is significant at 1% level of significance.
The Durbin-Watson statistic of 2. 079 signifies that autocorrelation is not present among independent variables. Table 3: Regression Model Summary R-Square 0. 244 F 9. 320 Durbin-Watson 2. 079 Table 4 gives the results of the regression. Out of the five factors analyzed, four factors viz. , leverage, liquidity, ownership structure and growth have expected relationships with the dividend payout. Whereas profitability, shows a sign contrary to what was expected. In line with literature, our results show that the leverage position of a firm has a negative relation with the dividend rate (-0.239), which is significant at 5% level of significance. Higher the exposure of the firm to external financing, higher will be the risk of the firm and therefore, lower would be the dividend payout. Table 4: Regression results Variables Dividend Rate (Constant) Expected sign Standardized Co-efficients Beta Dependent variable T Sig. 3. 840 .000 Leverage Negative -. 239 -3. 013 .003 Liquidity Positive .341 4. 138 .000 Profitability Positive -. 007 -0. 086 .932 Ownership Structure Positive .091 1. 113 .286 Growth Positive .041 .541 .589 Decision, Vol. 37, No. 2, August, 2010
The Determinants of Corporate Dividend Policy 71 Similarly, liquidity (. 341) shows a positive relation with the dividend rate at 1 % level of significance. The ownership structure of a firm representing institutional owners has a positive coefficient (0. 091) but is statistically not significant. Growth (. 041) also shows a positive coefficient, which is not significant. Contrary to what was predicted, profitability shows a negative coefficient (-. 007) but is statistically not significant. Thus, the results of our study indicate that there are two main determinants of dividend decision viz.leverage and liquidity. Conclusion Despite few decades of active research on a number of theories discussing determinants of corporate dividend policy, no significant judgment can be drawn. The present study reexamines the determinants of corporate dividend decision of Indian companies listed on the Bombay Stock Exchange during the period January 1, 2001–December 31, 2007. The study uses Principal Component Analysis for analyzing fifteen variables that have an impact on the dividend decision of a firm. The results gave five broad factors viz. , leverage, liquidity, profitability, ownership structure and growth.
These factors were then subjected to multiple regression with dividend rate as the dependent variable. The results of the regression show that leverage, liquidity, ownership structure and growth showed expected signs whereas profitability did not show the expected sign. Two factors viz. , leverage and liquidity were found to have a strong relationship with dividend rates of Indian companies. While leverage was found to be negatively associated, liquidity was positively related. One point worth mentioning here is that our results are drawn only from the analysis of financial factors affecting the dividend policy of an Indian company.
In practice some non-financial factors such as foreign collaborators’ shareholding, attitude and behaviour of management, company policies, etc may also have a bearing on the dividend decision of a firm. References Alli, K. L, Khan, A. Q. and Ramirez, G. G. 1993. Determinants of Corporate Dividend Policy: A Factorial Analysis. The Financial Review, 28(4): 523–547. Anand, M. 2002. Corporate Finance Practices in India: A Survey. Vikalpa, 27(4): 29 – 56. Anand, M. 2004. Factors Influencing Dividend Policy Decisions of Corporate India. The ICFAI Journal of Applied Finance, 2(10): 5 – 16.