The impact a dividend policy has on a firm

Dividend payout decision is a critical decision area and its one of the most important financial policies decision, not only from the viewpoint of the company, but also from that of the shareholders and others such as the employees and regulatory bodies. As the dividend literature suggests, if these decisions are handled efficiently, this is expected to be reflected in value of firms. More importantly, analysis of dividend policy is useful in enabling policymakers to identify the success or failure of policy initiatives or, alternatively, highlight different strategies undertaken by companies, which contribute to their successes.

Value of the corporate securities depends on dividend and, therefore, in deciding upon the financial structure of a company. Dividend decision in the corporate management is different from non-corporate entities. The corporate management is an elective

 one and the power of recommending dividend is vested with the board of directors. The board of directors therefore, decides the amount of dividend distributed by a company, and investors don’t have any direct say in such a decision. The questions of “Why do corporations pay dividends?” and “Why do shareholders, consumers, employees, government and regulatory bodies pay attention to dividends?” have puzzled both academicians and corporate managers for many years. Some researchers have developed and empirically tested various models to explain dividend behavior. Other researchers have surveyed corporate managers and institutional investors to determine their views about dividends. Despite much research intended to resolve the dividend puzzle, dividend policy remains one of the most judgmental decisions that a manager must make.

Dividend decisions may enhance the market value of the company but on the other hand it may mean less availability of internal funds and more dependence on external sources and expansion purposes. Furthermore, while determining dividend payment, a prudent management strikes a balance between shareholder’s expectation and firm’s long term interest. Dividend policy decision is affected by many factors. These factors may substantially vary from country to country. Dividend decision being an important financing decision of a company has attracted academicians and researchers in both developed as well as developing countries. But no study on the determinants of dividend payout policy has so far made in Bahrain. This has inspired the author to conduct this research, which is expected to guide the dividend policies of Bahraini companies. The objective of this paper is to contribute to the ongoing dividend debate to improve our understanding of corporate dividend policy, by providing evidence of dividend policies of Bahraini companies. More precisely this paper aims to identify the factors influencing cash dividend change in Bahraini companies listed in Bahrain Stock Exchange. Bahrain has an emerging stock market, the Bahrain Stock Exchange (BSE

) officially commenced operations in June 1989. Starting off with 29 listed companies back in 1989, this figure grew to 47 in early 2006, including seven non-Bahraini companies. The stock exchange’s market capitalization has risen from USD 2.7 billion in 1989 to USD 27 billion in 2007.

The remainder of this paper is structured as follows: section two reviews previous studies on the determinants of dividend policy and identifies the research hypotheses; the data and its analysis and research methodology is provided in section three; while section four presents the research results; finally, section five provides the conclusions.

2. REVIEW OF LITERATURE AND RESEARCH HYPOTHESES

2.1: Literature Review

Several theories have emerged to explain why firms pay dividends. These include agency cost reduction, clientele preference for dividend income, and signaling of current or future cash flows. A number of researchers have provided insights, theoretical as well as empirical, into the dividend policy puzzle. Lintner (1956) was one of the pioneers in studying corporate dividend behavior of 28 well-established industrial companies for the period of 1947-1953 by using a regression analysis and intensive interviews with managers responsible for the dividend decisions. Lintner found that major changes in earnings “out of line” with existing dividend rates were the most important determinant of a company’s dividend decisions, and concluded that a major portion of dividend of a company would be expressed in terms of company’s desired dividend payment and target payout ratio. However, because managers believed that shareholders prefer a steady stream of dividends, firms tended to make periodic partial adjustments toward a target payout ratio rather than dramatic changes in payout. Therefore, managers smooth dividends in the short run to avoid frequent changes.

Miller and Modigliani (1961), viewed dividends as irrelevant and they argue that given perfect capital markets, the payment of dividends does not affect the value of the firm and is therefore irrelevant. In such a world, firm value depends only on the distribution of future cash flows that result from the investments undertaken. Miller and Modigliani concluded that the value of company depends solely on its earnings power and is not influenced by the manner in which its earnings are split between dividends and retained earnings.

Researchers took different paths in identifying factors that influence dividend policy decisions. Some took a normative

 approach and developed theories about how companies should make dividend policy decisions. Others, who took a behavioral approach, went directly to managers and asked them what they actually consider when making such decisions. Comparing the responses from various surveys to the theoretical models provided a way of determining whether managers make dividend decisions in a manner consistent with academic theory (Baker et al, 2001). Most researchers and many academics greeted Miller and Modigliani conclusion with surprise because the conventional wisdom at the time suggested that a properly managed dividend policy had a significantly positive impact on share prices and shareholder wealth. For example, Bernstein (1996) stated that “… dividends, in and of themselves, do not matter, provided managers avoid driving down the spread between return on earnings and the cost of capital.”

Fama and Babiak (1968) studied the determinants of dividend payments by individual firms during 1947-1964. The authors used a regression model analysis, simulations and prediction statistical tests. Fama and Babiak found that net income seems to provide a better measure of dividend policy than either cash flow or net income and depreciation. Fama (1974) examined other models for explaining dividend behavior. Fama and Babiak (1968) and Fama (1974) results support Lintner’s view that managers prefer a stable dividend policy, and are reluctant to increase dividends to a level that cannot be sustained. Therefore, these researchers concluded that changes in per share dividends are largely a function of a target dividend payout based on earnings and the last period’s dividend payout.

Murray (1981) by using non-capital market data to test the theoretical implication that dividend payout is negatively correlated with earning uncertainty. Murray concluded that earning uncertainty is a determinant of the dividend policy. Baker et al. (1985) and Farrelly et al. (1986) surveyed 562 New York Stock Exchange (NYSE) firms. Based on their analysis of 318 responses from utility, manufacturing, and wholesale/retail firms, they found that the major determinants of dividend payments were the pattern of past dividends and the expected future earnings. The results also show that managers were highly concerned with dividend continuity and believed that dividend policy affects share value. Kim (1987) examined the effect of transaction costs and agency costs

 on dividend policy of 357 industrial firms during 1979-1981. The cross-sectional tests of the models related dividend payout ratios to explanatory

 variables such as past and expected future growth of firm, the total risk of firm and the number of stockholders. Kim concluded that transaction costs and agency costs effect firm’s dividend payout decision. Bond and Mougoue (1991) examined whether the speed of adjustment and target dividend payout rates implied in the partial adjustment model are an accurate characterization of corporate dividend policy. Bond and Mougoue conclude that the partial adjustment model does not generate unique measures of the dividend policy of the individual firm.

Pruitt and Gitman (1991) surveyed financial managers of the 1,000 largest US firms about the interplay among the investment, financing, and dividend decisions in their firms. Their results suggest that important influences on the amount of dividends paid are current and past years’ profits, the year-to-year variability of earnings, and the growth in earnings. Pruitt and Gitman also found that prior years’ dividends are an important influence on current dividends. Sharma and Rao (1992) tried to identify the signaling aspects of firm’s dividend payout decision. They concluded that dividends are perceived as signals from performance point of view, market’s point of view and management’s point of view. Karak (1993) examined the relationship between Indianan Company’s profits and dividend payout decision. Karak concluded that Indianan companies’ management in favor of financing expansions from internal resources also concludes that companies’ management followed conservative dividends policies. Simon (1994) studied the determinants of dividend payments by U. S. firms during two years 1984 and 1985. Simon reevaluated LintnerHYPERLINK “http://financial-dictionary.thefreedictionary.com/Lintner’s+Model”‘HYPERLINK “http://financial-dictionary.thefreedictionary.com/Lintner’s+Model”s model with new independent variables related to cash flows. His results support Lintner’s view that changes in per share dividends are largely a function of a target dividend payout based on earnings and the last period’s dividend payout. However, Simon found no relationship between cash flows and dividend policy.

Benartzi et al. (1997) by conducting a comprehensive they concluded that Lintner’s model of dividends remains the best description of the dividend setting process available. Baker et al. (2001) study reports the results of a 1999 survey 630 NASDAQ-listed firms. Based on their analysis of responses from 188 managers about the importance of 22 different factors that influence their dividend policy, they found that many managers of NASDAQ firms make dividend decisions consistent with Lintner’s (1956) survey results and model. Their results also show significant differences between the manager responses of financial and non-financial firms on nine of the 22 factors. Their results also suggest that managers pay careful attention to their choice of dividend policy for their firm. They conclude that because the dividend decision can affect firm value and, in turn, the wealth of stockholders, dividend policy is worthy of serious management attention.

Fama and French (2001) examine the characteristics of dividend paying companies. They find that three characteristics affect the decision to pay dividends: firm size, profitability, and investment opportunities. They note that larger firms and more profitable firms are more likely to pay dividends, whereas firms with more investment opportunities are less likely to pay dividends. Narasimhan and Vijayalakshmi (2002) analyze the influence of ownership structure of Indian firms on dividend payout and find no influence of insider ownership on dividend behavior of firms. Li, Feng, Song and Shu (2006) analyzed the decision-making of dividend policy and the reasons for dividends policy selection in non-state-owned listed companies in China by using structural equation modeling. The main research findings are as follows: (1) the dividend policy of non-state-owned listed companies in China can be interpreted by the western agency theory for dividend, and they found that if compared with manager, owner is a more important variable that influence the dividend policy, (2) four motives such as investment opportunities, refinancing ability, stock price and potential repayment capacity are all important factors for decision maker to determine the dividend policy.

Aivazian, and Booth (2003) find that for both U.S. firms and emerging market firms, profitability affects dividend payments, higher debt ratios correspond to lower dividend payments, and market-to-book ratio has a positive effect on dividend payments, in addition they find little evidence that business risk or size affects dividend payments. Ho, H. (2003) conducted a comparative study of dividend policies in Australia and Japan. They examines panel data the Australian stock market and the Japanese stock market. The results reveal that dividend policies in Australia and Japan are affected by different financial factors. Dividend policies are affected positively by size in Australia and liquidity in Japan, and negatively by risk in Japan only. An industry effect is found to be significant in both, countries.

Brunarski et al. (2004) examine the relation between agency costs and the firm’s decision to distribute excess cash to shareholders by declaring a nonrecurring special dividend or by significantly increasing the firm’s regular dividend. The evidence supports the notion that firms with greater agency costs are more likely to pay a special dividend, whereas firms with lower agency costs are more likely to increase their regular dividend. Dhanani ( 2005) by using a survey approach examines the importance and relevance of the various theories of dividend policy for UK companies, and evaluates the extent to which firm characteristics such as size and industry sector influence corporate managers’ views about the various dividend theories. The results in general support dividend hypotheses relating to signaling and ownership structure, in preference to those about capital structure and investment decisions and agency issues. At a more detailed level, the cross sectional analysis reveals important differences between managers’ responses, based on company size, industry sector, growth opportunities, ownership structure and information asymmetry.

Naceur et al (2006) examined the dividend policy of 48 firms listed on the Tunisian Stock Exchange during 1996-2002. They testes whether or not managers of Tunisian listed firms smooth their dividends. Also they highlight some determinants that may influence the dividend policy pattern. The results indicate that Tunisian managers, just like their counterparts in other emerging markets, do not smooth their dividend payments. Also results indicate that highly profitable firms with more stable earnings can afford larger free cash flows and thus pay out larger dividends, and fast-growing firms distribute larger dividends so as to appeal to investors.

Luciana and Aydin (2006) investigate the relationship between dividend policy and ownership structure of 139 listed Italian companies

. The results reveal that firms make lower dividend payouts as the voting rights

 of the largest shareholder increase. Results also suggest that the presence of agreements among large shareholders might explain the limited monitoring power of other ‘strong’ non-controlling shareholders. Pal and Goyal (2007) made empirical attempt to identify the leading factors that determine the dividend behavior of Indian banking industry. They found that lagged dividend, profit after tax, and interest are the most important factors affecting dividend decisions of Indian banking industry. Papadopoulos and Charalambidis (2007) investigate the present status and determinants of dividend policy of 72 companies listed in the Athens Stock Exchange

 during 19952002. Their findings are summarized as follows: 1) the proportion of firms distributing regular and total dividends is almost stable across the eight-year period; 2) most Athens firms distribute no special dividends; 3) differences between dividend policy of retail firms and that of industrial firms are minor; 4) the variables used explain only a small proportion of dividend policy’s variability; and 5) cash flow is the most important dividend payout determinant and is positively related to the proportion of earnings distributed either as a regular or as total dividend.

2.2: Research Hypotheses

The primary purpose of this paper is to identify the factors influencing cash dividend change in companies listed in Bahrain Stock Exchange. Finance scholars have engaged in extensive theorizing about factors that may be important in determining a firm’s dividend policy. For example, some of these theories involve tax preference, signaling, and agency explanations (see for example: Rozeff, 1982; Easterbrook, 1984; Farrelly and Baker, 1989; Barclay et al, 1995, Baker et al 2000; DeAngelo, H. et al., 2000; and Baker et al 2001). According to some researchers, dividends may increase when earnings increase. Company profitability may have signaling implications for well performing firms in its capacity as an indicator of future performance. To the extent that profitability and dividends serve as substitute indicators, high profitability firms are less likely to look to dividends as an important signaling mechanism (Easterbrook, 1984). On the other hand, if dividends and profitability are complementary, dividend relevance for these firms will increase.

Other researchers assume that size of firms is negatively related to both agency conflicts and information asymmetries. Small firms face greater information asymmetry than large firms and are consequently more likely to use dividends as a signaling mechanism (Mougoue and Rao, 2003). Moreover, dividend policy may be a more amenable signaling tool for smaller firms in comparison to alternative communication methods. At the same time, however, smaller firms may face more severe conditions to raise capital externally and are consequently more likely to follow a dividend preservation strategy (Ooi, 2001). DeAngelo and DeAngelo, (1990) empirical study findings indicate that dividend distribution is negatively related to firm’s size.

This study addresses the following research question: What are the most important determinants of dividend payout policy of Bahraini firms? Based on previous studies we advance the following four hypotheses in response to that question:

H1: Profitability affects Bahraini firm’s dividend payout policy decisions.

H2: The pattern or continuity of past dividends affects Bahraini firm’s dividend payout policy decisions.

H3: Financial leverage affects Bahraini firm’s dividend payout policy decisions.

H4: Size of firm affects Bahraini firm’s dividend payout policy decisions.

3. RESEARCH DATA AND METHODOLOGY

3.1: Sample Data and Study Period:

To test the hypotheses above, a sample of Bahraini listed companies of the Bahrain Stock Exchange during 2006 and 2007 was collected. Data needed data to accomplish this study has been extracted from the annual reports of Bahraini companies. These annual reports represent the most recent source of data available at the time of the study. Also, data has been extracted from “Bahrain Stock Exchange Investors’ Guide (2007)” which published by Bahrain Stock Exchange.

The study sample consist of 35 companies were selected from 47 companies. Inclusion in the study sample required that companies meet the following screening criteria: (1) the companies had to be Bahraini corporation that were listed on the Bahrain Stock Exchange during 2006 and 2007; (2) the company had to have paid a cash dividend in at least one year during 2006-2007 period; (3) availability of complete information on all variables. These criteria used to have a sample consisting of Bahraini corporations that have recently paid cash dividends. Using these criteria, the resulting sample contained 35 Bahraini companies covered six sectors commercial banks, investment, insurance, service, hotel & tourism, and industrial companies. The companies understudy have been represented by 6 banks, 12 investment companies, 4 insurance companies, 8 service firms, 3 hotel & tourism companies, and 2 industrial firms. Of these companies, 3 companies paid cash dividends during only one of the years, while the remaining 32 companies paid cash dividends during 2006 and 2007. The period of study covers two years 2006 and 2007.

3.2: Independent variables:

The choice of variables included in or omitted from a regression model and the definition used in the estimation of important factors can significantly influence a study’s results (Short, 2002), therefore, one of the most difficult issues in empirical dividend payment policy studies is the determination and the definition of independent variables. The independent variables used in the current study are as follows: profitability, cash dividends of pervious year, financial leverage and size of firm.

The independent variables were defined or measured as follows:

1. Profitability is measured by dividing the companies’ net income of year 2007 over companies’ market capitalization of year 2006.

2. Cash Dividends of pervious year is measured by dividing the companies’ cash dividends of year 2006 over companies’ market capitalization of year 2006.

3. Financial Leverage is measured by dividing the book value of total debt to the book value of ownersHYPERLINK “http://financial-dictionary.thefreedictionary.com/owners’+equity”‘HYPERLINK “http://financial-dictionary.thefreedictionary.com/owners’+equity” equity, and then companies’ financial leverage of year 2007 divided over companies’ market capitalization of year 2006.

4. Size of firm is measured by dividing the companies’ total assets of year 2007 over companies’ market capitalization of year 2006.

3.3: Dependent variable:

Change in cash dividends represents the dependent variable in the regression model. Change in cash dividends is measured as follows: first of all the different between cash dividends of year 2007 and cash dividends of year 2006 were computed, and then the companies’ cash dividends different divided over companies’ market capitalization of year 2006.

3.4: Model Development:

The following regression model was fitted to the data in order to assess the effect of each independent variable on the dependent variable (i e change in cash dividends).

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

[DELTA][DIV.sub.it]/[MV.sub.i(t-1)] = Change in Cash Dividends for firm i in period t;

[NI.sub.it]/[MV.sub.i(t-1)] = Net Income for firm i in period t;

[DIV.sub.i(t-1)]/[MV.sub.i(t-1)] = Cash Dividends for firm i in period t-1;

[FL.sub.it]/[MV.sub.i(t-1)] = Financial leverage for firm i in period t;

[TA.sub.it]/[MV.sub.i(t-1)] = Total Assets for firm i in period t;

e = error term.

4. RESULTS

Table 1 show the descriptive statistical tests results of independent variables for the study sample of companies. The table presents the minimum, maximum, mean, standard deviation, and variance for all independent variables in the regression model.

Table 2 presents correlations coefficients between all variables. The results show there is some moderately high correlations between variables, more specifically between Chang in Dividends and net income, and Chang in Dividends and total assets, and between net income and dividends in prior year. However, it has been suggested (Farrar and Glauber, 1967; Judge et al., 1985) that correlation coefficients should not be considered harmful until they exceed 0.80. Table 2 reveals that the highest correlation is (0.708) between Chang in Dividends and net income. Therefore, collinearity did not appear to be a serious problem in interpreting our regression results.

The [R.sub.2] and Adj [R.sub.2] and F-value for the model are presented in tables 3 and 4. The adjusted coefficient of determination ([R.sub.2]) indicates that 68.7% of the variation in the dependent variable is explained by variations in the independent variables. Table 5 presents a summary of the regression model results for the change in dividends. Standardized beta coefficients, t-statistics, and probability levels are given for each independent variable.

The regression model results in Table 5 indicate that there is a very strong positive relationship between change in dividends and profitability (Net Income) and it’s statistically significant at 1% level. This result is consistent with that found in many previous empirical studies (e.g. Lintner, 1956; Fama and Babiak, 1968; Baker et al., 1985; Farrelly et al., 1986; Simon, 1994; Fama and French, 2001; Short et al., 2002; Naceur et al.,2006; and Pal and Goyal, 2007), this result suggest that companies’ profitability is the major determinant of Bahraini companies’ cash dividend decision. This evidence supports the first hypothesis.

Table 5 regression results also indicate that the dividend in prior period (Dividend t-1) factor is statistically significant at 1% level. This evidence supports the second hypothesis. This result suggests that there is a very strong negative relationship between prior period cash dividend in Bahraini companies and their dividends policy. This finding is consistent with that found in many other previous empirical studies (e.g. Lintner, 1956; Fama and Babiak, 1968; Fama 1974; Baker et al., 1985; Farrelly et al., 1986; Kim, 1987; Pruitt and Gitman, 1991; Benartzi et al., 1997; and Baker et al., 2001) which suggest that the pattern or continuity of past dividends is one of the major determinants of dividend payments, also suggest that managers are highly concerned with cash dividend continuity and believed that dividend policy affects share value of the company.

Moreover, the regression model results in Table 5 also indicate that there is a strong negative relationship between change in dividends and size of firm (Total Assets) and it’s statistically significant at 3% level. This evidence supports the fourth hypothesis. This finding is consistent with that found in other previous empirical studies (e.g. DeAngelo and DeAngelo, 1990; Fama and French, 2001; and Aivazian, and Booth, 2003) which suggest that the size of companies listed in Bahrain Stock Exchange is one of the main determinants of Bahraini firm’s cash dividend payout decision. The negative relationship between change in dividends and size of firm is consistent with the assumption of (Mougoue and Rao, 2003) that, size of firms is negatively related to both agency conflicts and information asymmetries. Small firms face greater information asymmetry than large firms and are consequently more likely to use dividends as a signaling mechanism. However, the regression results show no relationship between change in dividends and financial leverage.

5. CONCLUSIONS

Dividend payout policy continues to be an often conversed area between financial economist and corporate managers. Many researchers have worked on dividends and proposed several theories to explain dividend behavior. However, the dividend set remains puzzling. This paper resides in providing a further insight into dividend payout policy within the Bahraini emerging capital market. The aim of this paper is to identify factors influencing cash dividend change in companies listed in Bahrain Stock Exchange. A sample of (35) Bahraini companies were tested. Descriptive and statistical tests were used to analyze the data of the study. Cash dividend change in Bahraini companies is tested with regard to four specific characteristics, namely: profitability; previous year dividends; financial leverage and size of firm. The regression model results indicate that cash dividends policy is significantly related to profitability, change in previous year dividends, and size of Bahraini companies listed in Bahrain Stock Exchange, but not to financial leverage.