The Impact Of Dividend Policy On Share Price

Published Date: 02 Nov 2017

The impact of dividend policy on the stock price of a firm is still a controversial issue in the field of finance over last few decades. Many researchers and economists have engaged in analyzing and testing company dividend policy since many years ago. However, there are still continuing debates on this issue until today. The harder we look at the dividend picture, the more it seems like a puzzle, with pieces that do not fit together. (Black, 1976)

Dividend is a portion of a firm’s current or retained earnings distributed to its shareholders. It can be in the form of cash or share of the firm. Dividend policy is commonly defined as a policy implemented by the firm to decide how much dividend it will pay out to its shareholders. The management has to decide whether to pay out or retain the profits once the firm makes profit. The firm may implement a dividend policy which may impact the investors and future prospects of the firm in the financial markets once the firm decides to pay out dividends to its shareholders. The decision depends on the current situation of the firm and also the preferences of investors. The main objective of the management is to maximize the shareholders’ wealth. Therefore, this objective can be achieved by paying an appropriate amount of dividend to the shareholders on their investments. However, the impact of dividend decisions on shareholders’ wealth is still a debatable issue.

In general, the market price of a firm’s stock represents the shareholders’ wealth which is important for the firm’s investment, financing and dividend decisions. In a sense, the dividend decision is essentially important for the achievement of the firm’s objectives and efficient performance because the role of the finances is increasingly significant in the firm’s growth strategy. Therefore, management of the company has to determine an appropriate dividend policy in order to increase the value of the firm. There is an argument that a decrease in the dividend payment will reduce the market share price of the firm.

There are several types of dividend policy such as stable dividend policy, constant dividend policy and residual dividend policy. Stable dividend policy offers a stable dividend payout ratio. It has a positive impact on the firm’s share price. It includes constant dividend per share and constant payout ratio. The management will not simply decrease the dividend until the company is incapable to support the present dividend. On the other hand, residual dividend policy states that the dividend should be paid only out of left over earnings. This is because the investors prefer the firm to retain the money for further investment if the return on the firm’s investment exceeds that of the individual investor investment of a similar risk. The company can also save on issue cost such as legal fee, underwriting fee and underpricing.

Besides that, Lintner (1956) found that companies have long run target dividend payout ratio which can only be achieved in the long run. Management also focuses more on dividend changes rather than on absolute value. Besides that, he also found that managers smooth dividend. Therefore, temporary changes in earnings are unlikely to lead to changes in dividend as managers do not like erratic dividend payment due to uncertainty. Furthermore, managers are reluctant to make dividend changes that might have to be reversed as this will lead to uncertainty among shareholders.

On the other hand, Miller and Modigliani (1961) argue that dividend is irrelevant and does not affect the firm’s value under perfect capital market. The reason is that there is no agency cost, asymmetric information, tax, transaction cost and flotation cost in perfect capital market.

Many past studies state that dividend is a signaling tool. This is known as dividend signaling theory. It indicates that high dividend payment provides a positive signal about the current and future prosperity of a company to outsiders. This is in line with the efficient market hypothesis where all information available to investors is quickly and accurately incorporated into the share price. However, some researchers found no evidence of that. Therefore, another dividend issue still remains unsolved.

Moreover, dividend can be used to reduce agency problem. Agency theory deals with the potential conflict of interest between shareholders and management. Dividend payment will reduce the amount of free cash flow within the company. Therefore, it reduces the ability of manager to waste shareholders’ funds on questionable acquisition. As a consequence, dividend payment is relevant as it reduces agency cost and indirectly increases the firm value. (Jensen and Warner, 1988)

In Malaysia, firms are free to decide how much they will pay out in dividend to their shareholders for a specified business year in accordance with Companies Act 1965. Shareholders prefer cash dividend and also growth in earning per share (EPS). Firms should implement optimal dividend policy that maximizes the market share price of the firm and therefore improves economic growth. The study highlights the relationship between dividend payout and market share price of a firm. It also examine whether there is impact of dividend policy on shareholders’ wealth.

Problem Statement

The effect of dividend policy on share price is still a controversial topic for many years. Dividend refers to a portion of a firm’s current or retained earnings distributed to its shareholders. Whether it has impact on the market share price of a firm is still questionable. None of researchers and financial analysts can actually determine the best dividend policy. Some researchers believe that dividend payment will increase the firm value. However, some believe that dividend payment will decrease the firm value instead. Furthermore, some researchers also emphasize that dividend payment has no impact on the firm value. For instance, Miller &Modigliani (1961) states that dividend payment and share price are irrelevant under perfect capital market where there is no tax, transaction cost and asymmetric information.

Furthermore, leftist states that a firm will pay less or no dividend if dividend is taxed more heavily than capital gain. The money should be retained within the firm to finance new investment for future capital gain (Myers, Brealey and Allen, 2008). Conversely, Lintner (1956) found that investors prefer high dividend policy as dividend in hand is less risky than capital gain. Besides that, Watts (1973) found that a company’s current and past dividends did have a little effect on the company’s future prospects. As a result, managers are difficult to determine which dividend policy is best fit to the company.

Why managers still intend to pay dividend to the company’s shareholders if dividend is irrelevant to the share price? If dividend does have an impact on the firm’s value, which dividend policy theory can best describe the relationship between dividend payout and share price? Therefore, this study wants to explain whether cash dividend payment and other stated factors (earnings per share, firm size and debt ratio) and share price are relevant in selected Malaysian companies.

Objectives

Objectives of this research are:

To examine the relationship between dividend payment and the share price of the selected companies in Malaysia.

To determine which dividend policy theory corresponds with Malaysian companies’ dividend phenomenon.

To identify whether other factors such as earnings per share, firm size and debt ratio play a role in the impact of share price of selected Malaysian companies.

Hypotheses

• Hypothesis 1:

H0: There is no significant statistical relationship between cash dividend and the market share price of companies in Malaysia.

H1: There is a significant statistical relationship between cash dividend and the market share price of companies in Malaysia.

• Hypothesis 2:

H0: There is no significant statistical relationship between earnings per share (EPS) and the market share price of companies in Malaysia.

H1: There is a significant statistical relationship between earnings per share (EPS) and the market share price of companies in Malaysia.

• Hypothesis 3:

H0: There is no significant statistical relationship between firm size and the market share price of companies in Malaysia.

H1: There is a significant statistical relationship between firm size and the market share price of companies in Malaysia.

• Hypothesis 4:

H0: There is no significant statistical relationship between debt ratio and the market share price of companies in Malaysia.

H1: There is a significant statistical relationship between debt ratio and the market share price of companies in Malaysia.

Research Design

The deductive approach is used to formulate hypotheses and measure them empirically. Quantitative approach relies on numerical evidence to form conclusions or to test hypothesis. Analytical designs are used in this research for the analysis of the relationship between dividend payments and market share price of Malaysian companies to achieve the objectives of this research. Explanatory method is also applied in this research to explain the relationship between the four chosen factors and market value of share of those selected Malaysian companies. Explanatory method is used as it could recognize and meet the purpose of explaining the relationship between the independent variables and the dependent variable.

Data collection

103 companies in Malaysia from several industries are chosen as the sample. The requirement is that companies should have paid cash dividend for the past five financial years (year 2008-year 2012). Secondary sources are collected from library, reference books and journals, corporation financial statements and audit reports which similarly conducted by other researchers. These secondary data are used for references and aid in research writing.

Resources

Library and its resources - reference books and articles in the library which are enrich with information assisting research findings.

Computer and its software - Laptop is used for data storage, writing purpose using Microsoft words and presentation purpose using Power Point. Statistical Package for Social Sciences (SPSS) program is used to analyze and test the hypotheses by calculating the means, standard deviations, correlation coefficient.

Internet - Internet is significant for online information gathering using search engines such as Google or EBSCO host and communication between group members via electronic-mail.

2.0 Literature Review

2.1 The Impact of Various Dividend Policies on Share Price

2.1.1 Irrelevant Dividend Policy

Miller and Modigliani (1961) concluded that dividend policy has no impact on a company’s share price under perfect capital market, that is, with no agency cost, tax, asymmetric information, transaction cost and floatation cost. The original shareholders transfer value to new shareholders when the firm issues new shares to finance the dividend. They claimed that only the firm’s investment policy will affect its share price. As a result, dividend policy will not help managers to maximize shareholders’ wealth.

Black and Scholes (1974) tested the relationship between share return and dividend yield based on 25 portfolios of ordinary shares listed on NYSE for the period from 1936 to 1966. The purpose was to determine the impact of dividend policy on market share price. They applied capital asset pricing model (CAPM) for testing the relationship between dividend yield and expected share return. Their results found that the coefficient of dividend yield was not significantly different from zero. Their research empirically supported the theory and concluded that the returns for investors who are holding low or high yield shares are the same.

However, Ball and Brown (1979) concluded that the dividend policy is relevant since he tested the impact of dividend on the share price based on Australian data from the period of 1960 to 1969 by applying the cross-sectional regressions model. He stated that the no-retention assumption created by M&M does not make sense as it is impossible for a firm to distribute all its free cash flow to its shareholders and thus each firm will retain some profits for further investment. As a result, the dividend policy is said to be relevant because suboptimal policies could be chosen by managers to invest in positive NPV projects. Besides that, Baker and Powell (2000) interviewed 603 listed US firms’ CFOs by designing questionnaire and did a survey about their behaviors. The survey proves that ninety percent of interviewers did not agree with M&M’s irrelevant dividend policy theory. It also showed that 87% of respondents agree with the interrelationship between dividend and the firm’s financing and investment policies.

2.1.2 Dividend Policy based on Bird-In-Hand Theory

Gordon conducted this study in 1959 and concluded that investors prefer dividend payments to capital gains as dividends are safer and they provide investors a secure income. He applied a two-year (1951 and 1954) sample data of four industries including steels, food, chemical and manufacturing tools industries. He then tested them by analyzing various regression models based on the data. The dividend hypothesis was examined by using linear regression model. The results proved that future capital gains received by investors will have higher fluctuation than dividends today. According to his study, higher dividend payout will cause investors to discount a higher dividend stream at a lower required rate of return. He also argued that a firm’s shares with high dividend payout normally have a lower risk. In a sense, less risky shares will normally have a higher price given that other factors of determining share price remain unchanged. As a result, this combined effect will then increase the firm’s share price and its value as the risk (ke) will be lower when the firm distributes dividend to shareholders.

Diamond (1967) reexamined Gordon model and corrected his regressions model equation by changing R to a three-year mean value centered on t-1. He tested the effect of retained profits and dividends on stock prices based on 255 listed US companies from 8 industries for the year of 1961 and 1962. His results showed that high growth industries prefer more retained profits than dividends whereas low growth industries prefer dividends to retained profits. As a result, he stated that the growth of company associated negatively with the dividend payout. His outcome was quite similar to early results of Friend and Puckett (1964).

Irfan and Nishat (2003) concluded that there is a positive relationship between dividend payout ratio and the volatility of the share price based on their research on 160 Pakistani companies for the period from year 1981 to year 2000 based on cross sectional regressions model. However, Adeoti, Oladipo and Adefila (2004) stated that investors of Nigerian companies prefer the company distribute the dividends to them based on their expectations. By using Pearson Product Moment correlation model, they claimed that there is no relationship between dividend payout, net profit and the company’s value as company distributes dividend to satisfy its shareholders regardless of its profit earned.

Moreover, Bajer, Powell and Veit.E.T, (2002) interviewed NASDAQ companies’ managers to examine the Bird-in-Hand effect on market share price. Their results found that 17.2% of 186 responses prefer dividends to future capital gains, 28% of them remained neutral, and 54.9% of them rejected the Bird-in-Hand theory. They then concluded that their study did not empirically give evidence to support the bird-in-hand definition.

2.1.3 Dividend Policy based on Agency Theory

Agency theory deals with the potential conflict of interest between shareholders and management. Conflicts happen when managers ignore shareholders’ interests and focus only on their personal goals. As a result, shareholders may want to control managers’ behaviors through some policies. Monitoring managers will bring high costs to a company. Generally, shareholders may find it difficult to control managers’ actions as tasks and decisions made by managers are most likely difficult to be understood by shareholders. Therefore, a specialized outsider plays an important role for monitoring them. For instance, managers can be monitored effectively through external financing. In brief, dividend payment will decrease the amount of free cash flow within a company. Therefore, shareholders’ fund will not be simply wasted by managers on questionable acquisitions since the reduction of cash will negatively affect managers’ performance. Managers will then issue equity to finance the investments once they get approval from Security Commission. In a sense, Security Commission is aiding shareholders to monitor managers’ behaviors. As a result, dividend payment is said to be relevant since it helps lower agency cost and indirectly increases the firm value and its market share price.

Furthermore, Rozeff (1982) found that dividend policy will decrease managers’ cash in hand and thus it leads them to borrow money from outsider to finance the firm’s investment. This can help shareholders to control managers’ actions through extra control of external borrower. He developed a model that underpins this theory, called the cost minimization model. The model combines the transaction costs that may be controlled by limiting the payout ratio, with the agency costs that may be controlled by raising the payout ratio. Indeed results from an Ordinary Least Squares cross sectional regression using 1981 data on 1000 US firms, support the theory put forward. Thus the model provides good fit and consequently has attracted the attention of subsequent studies.

Roll (1998) found that it is not profitable for a company to perform merger as the price for merger is most likely higher than the market value. However, managers tend to purchase other companies for their desires to raise scope of control within the company. As a result, this agency cost can be eliminated by simply paying higher dividends in order to avoid over-investment by managers. Furthermore, Llyod, Jahera and Page (1985) modified Rozeff’s cost minimization model by adding a size variable. An OLSQ cross sectional regression is applied to 1984 data on 957 US firms, and the results provide support for the cost minimization model and show that firm size is an important explanatory variable.

Pettit (1976) observed that managers normally know more about the firm’s performance than its shareholders due to asymmetric information. However, managers may not be able to disclose information to the firm’s shareholders due to reasons like difficulties to prove the information or for the fear of being sued by Security Commission. Therefore, managers will use other method like paying higher dividend to provide positive signal to the company’s shareholders. As a result, the information gap between shareholders and managers will be smaller.

2.1.4 Dividend Policy based on Signaling Theory

Asquith and Mullins (1983) tested the market response to dividend announcements based on sample of 168 companies that pay dividend for the first time or continued distributing dividends after ten-year hiatus. They examined the average daily excess share returns ten days surrounding the notice of dividend initiation. Their results showed that positive 3.7 percent of excess return was found for the period of two-day announcement. Furthermore, they stated that there was a significant relationship between the importance of initial dividend and excess share returns on the day of announcement by applying the cross-sectional regression.

Watts (1973) carried forward the results of Pettit (1972). In his model, it investigated the importance of dividend in two different ways. Firstly the relation whether the unexpected dividend changes were associated with positive future earning changes and secondly the relation between unexpected changes in share price and changes in annual dividends which convey priceless information to shareholders. The sample consisted of 310 firms. The results showed that relationship was positive but not very strong.

Amihud and Li (2006) provided evidence for the statement that modification of dividend communicates information about companies’ values. They applied 255 dividend increase events and 51 dividend decrease events from year 1988 to year 1992 to test the impact of announcement on share price in US. In their model, they base their argument on the declining dividend response coefficients (DRCs) in the face of increasing institutional holdings. Their results showed that there were dividend increase of +0.965% and dividend decrease of -1.73% for average excess stock return.

Furthermore, Travlos (2001) also gave supports that signaling theory plays a significant role in emerging market. 41 cash dividend increase announcements and 39 share dividend announcements from year 1985 to year 1995 were applied as sample in the study. The objective was to test the market response based on those announcements. The outcome showed positive and significant relations and thus the signaling theory was supported by the study.

However, Amihud and Li (2006) observing this theory provided inconsistent results. Koch (1996) argued that dividends increased by the company signal its past but not its future.

2.1.5 Dividend Policy based on Tax Differential Theory

Brennam (1970) stated that a company will not distribute dividend to its shareholders whenever the dividend is taxed more heavily than the capital gain. Shareholders prefer the firm to retain the profits rather than distribute it to them. These undistributed profits will then be used by the firm to finance its new investments. Therefore, shareholders should invest more in those shares with low or no dividend for future capital gain. He derived the after-tax capital asset pricing model (CAPM), where the before-tax return of stocks is positively related to the tax burden of equity securities in the cross section.

However, Black and Scholes (1974) examined Brennam’s model and concluded that the tax effect is not relevant. They argued that the capital gain is not affected by the dividend yields either before or after tax. Furthermore, Litzenberger and Ramaswamy (1982) disagreed with Black and Scholes’s result. They criticized their classification of dividend yield. They then applied monthly dividend yield classified into positive yield shares and zero dividend yield shares to prove Brennam model. Their results found that the dividend coefficient was 0.236 and it was highly significant and concluded that Brennam model is relevant.

Keim (1985) applied Sharpe CAPM model to examine the relationship between long term dividend yield and share returns. 429 companies in US were used as a sample in his research from the period of 1931 to 1978. He formed six dividend-yield portfolios that consisted of positive dividend-yield and zero dividend-yield companies. His outcomes recommended yield-related tax effect.

2.1.6 Impact of Dividend Policy on Firm Risk

Rozeff (1982) did a research on the determinants of dividend payout ratios. He recommended that growth, agency cost and beta play a role in influencing the optimal dividend payout. He suggested that higher beta coefficients will lead to lower dividend payout. He used the multiple regressions model to test the association of dividend policy with the firm risk with the sample size of 1000 firms from sixty four different industries. It was found that there is significant negative relationship between corporation risk and the dividend payout. He then justified that firms with high beta normally distribute lower dividend because of the higher external borrowing cost.

Ben-Zion and Shalit (1975) examined how firms’ size, leverage and dividend influence the risk of ordinary stock. 1000 largest US industrial companies in 1970 were chosen as the sample. Ordinary Least Square regressions model was applied in the study to examine their relationships. They then tested whether there is a significant relationship between alternative risk and firms’ size, leverage and dividend payment by using the correlation matrix. The findings of the study concluded that the firm’s size, leverage and dividend payment have a significant relationship with the corporation risk measures. Besides that, they are listed as the important determinants of firm risk. They showed that the corporations’ risk has a significant negative relation with both the firm size and the dividend yield. However, the corporation risk is negatively affected by the leverage.

Eades (1982) examined the relationship between corporation risk and dividend yield based on US stock market. Firms related to 20 years (1960-1979) annual COMPUSTAT files and the monthly share return files from Centre for Research in Security Prices (CRPS) are used as the sample. He applied the simple regression model for the first set of coefficient and used multiple regressions model to obtain the second set of estimates results. Furthermore, he regressed the dividend yield against the beta. The findings stated that there is a significant negative relationship between dividend yield and beta. It concluded that firms with high beta may distribute lower dividend.

2.1.7 Relationship between dividend policy and share price volatility

Baskin (1989) tested the relationship between dividend policy and share price volatility by applying a different method. Some control variables are added in order to examine the relationship between the dividend policy and the share price volatility. Those control variables consisted of growth, debt, earning volatility and the firm size. These variables clearly affected the share price volatility. They also influence the dividend yield. For example, the earning volatility not only affects the stock price volatility but also influence the firm’s optimal dividend policy. Besides that, assume that the operating risk is unchanged, the debt level may positively affect dividend yield. The firm size may also have an effect on the stock price volatility. Large corporations are mostly more diversified and thus the value of large corporations might be more stable than those of small corporations. Moreover, investors may react irrationally due to the limited public information from small corporations.

Baskin applied the analytical models that link the dividend to the risk of share to his study. The models consisted of information effect, rate of return effect, arbitrage pricing effect and duration effect. The duration effect stated that fluctuation in interest rate has less effect on high dividend yield shares. Corporations that have high dividend yield will have less volatility in stock price since high dividend yield is likely said to be a positive signal of the cash flow. He then applied the Gordon growth model to prove this effect. Furthermore, the rate of return effect by Baskin stated that corporations with low dividend payout and dividend yield can be evaluated more valuable than their assets because of the growth opportunities. By referring to his study, corporations with low dividend payout and dividend yield will have more fluctuation in stock price because the predictions of profit from growth opportunity have more error than the forecast of profit from their assets. He also stated that the excess return is subordinate of price discount rate and the dividend yield since the arbitrage profit will increase by having higher dividend yield. The study also concluded that the share price volatility and the risk of share normally can be controlled by managers when setting the dividend policy. The announcement of dividend distribution may reflect the stability of corporations.

2344 US companies were used as the sample in the study and the time period was from 1967 to 1986. The study concluded that there is a significant negative relationship between the dividend yield and the share price volatility. Besides that, there is also a significant relationship between other variables and the stock price volatility. As a result, Baskin stated that the stock price volatility can be controlled by a corporation’s dividend policy. He also concluded that a decrease in standard deviation of share price trend will lead to an increase in dividend yield. For instance, if the dividend yield rises by 1%, the standard deviation of share price will decrease by 2.5% as stated in the study.

Nazir et al (2010) examined the relationship between dividend policy and stock price volatility based on the sample of 73 companies listed in Karachi Stock Exchange (KSE) from the period of year 2003 to year 2008. The panel data was tested by using the fixed effect and random effect models. The result was that there is a significant negative relationship between dividend payout and dividend yield and the stock price volatility. Besides that, the study also concluded that there is non-significant negative relationship between leverage and firm size and the stock price volatility.

Suleman et al (2011) examined the relationship between dividend policy and stock price volatility based on the sample of Pakistan companies from five different sectors from year 2005 to year 2009. The data was collected from Karachi Stock Exchange (KSE). They analyze the data by applying multiple regressions model in order to test their associations. The findings stated that there is a significant positive relationship between the dividend policy and the stock price volatility. The result was opposed to Baskin’s findings. The result also stated that the company’s growth affects the stock price volatility negatively.

Hussainey (2011) studied the association of stock price volatility with the dividend policy based on UK firms. 123 corporations were chosen as the sample and the time period was from year 1998 to year 2007. They applied the multiple regressions model to examine the association between dividend payout and dividend yield and the stock price. Several control variables were added to their model in the study such as earning volatility, firm size, growth rate and debt level. The findings stated that there is a significant relationship between the dividend payout and the stock price volatility. The result was similar to the findings of Allen & Rachim (1996)’s study. Hussainey concluded that there is a negative relationship between the dividend yield and the stock price volatility. As a result, the dividend payout ratio is stated as a major determinant of the stock price volatility according to the study. Besides that, the debt level and the firm size affect the share price volatility the most among the control variables. The findings stated that the firm size affects the share price volatility negatively which was opposed to Allen & Rachim’s findings. However, the stock price volatility is affected positively by the level of debt according to the research.

2.1.8 Impact of Dividend Policy on Firm’s Performance

Tian et al (2006) examined the relationship between the dividend policy and the company performance. 406 corporations were used as the sample from the period of year 1989 to year 2000. The results showed that there is a significant relationship between the dividend policy and the company performance. The study applied the model of autocorrelation test in order to examine the relationships. According to the result, changes in dividend strongly connect to current and past performance since positive coefficient of changes in dividend was found after the performance announcement.

Furthermore, Amidu (2007) also examined the relationship between the dividend policy and the company performance. Twenty five corporations listed on Ghana Stock Exchange (GSE) were chosen as the sample and the time period of the study was from year 1997 to year 2004. Data was collected from those selected firms’ annual reports. In the study, pooled panel crossed-section regression model was applied for the purpose of getting the maximum possible observation. The panel regression model was not similar to the regular cross section or time series regression model since the subscript was doubled to each variable (dividend policy, dividend per share, firm size, debt ratio and growth in sales). The findings showed that there is a significant positive relationship between the dividend policy and return on assets and sales growth. As a result, the research also supports the second school of thought that the company performance is affected significantly by the dividend policy.

Moreover, Azhagaiah and Priya (2008) concluded that high dividend payment will lead to an increase in market value of stock. Twenty eight corporations from Indian chemical industry were selected as the sample and the time period of the study was from year 1996 to year 2006. By applying stepwise regression model and multiple regressions model in the study, five variables were tested whether they have significant effect on the shareholders’ wealth. These variables consisted of cost of capital, improvement of profit margin, capital structure decision, sales growth and capital investment decisions. The findings stated that there is a significant relationship between the dividend policy and the stock price in those Indian chemical corporations. Therefore, it can be explained that the dividend policy plays an important role to decide the shareholders’ wealth and thus shareholders prefer current dividend to future income.

Chapter 3: Theoretical Framework and Research Methodology¬

3.1 Theoretical Framework

Many past studies have recommended various factors that would affect a company’s share price. In this study, cash dividend will be chosen as the main factor that would affect the firm value. However, there are other factors that could affect the share price as well. Thus, another three factors that were suggested in the past studies most frequently will be selected as control variables in this study. They are net earnings per share (EPS), firm size (total asset) and debt ratio.

The main objective of this study is to examine whether there is a significant relationship between these four factors and company’s share price of selected companies in Malaysia.

Independent Variables

Dependent Variable

Figure 3.1: Theoretical Framework

3.2 Dependent Variable

The dependent variable in the study is the market value of company’s share. Each year’s closing share price will be used as the proxy for this variable.

3.3 Independent Variables

Dividend Policy: The amount of cash dividend (dividend per share) of a company each year is applied as the proxy for this variable.

3.4 Control Variables

Three control variables are applied for this research. These variables are chosen as they are recommended in past studies frequently. They are expected to have a great effect on the market value of share and listed as following:

a) Profit: Earning per share (EPS) for each year of a company is applied as a proxy for this variable. Baskin (1989), Allen & Rachim (1996), Liu & Hu (2005), Adefila, Oladipo & Adeoti (2004), Adesola & Okwong (2009) and Chen, Huang, & Cheng (2009), Khan, Aamir, Qayyum, Nasir, & Khan (2011) also used Earnings per Share as a control variable in their study and report the positive relation between Earnings per Share and Stock Prices while Adefila, Oladipo & Adeoti (2004) does not find any relation between Stock Prices and Earnings.

b) Firm size: The amount of total asset at the end of each year is applied as a proxy for this variable. The firm size is one of the control variable measured by using the natural logarithm of total asset (Smith and Watts, 1992; Kouki and Guizani, 2009; Chae et al., 2009).

c) Debt ratio: The firm’s debt to equity ratio is applied as the proxy for this variable as recommended by most of the past studies such as Lev and Kunitzky (1974), Gaver, and Gaver, (1993), Gul, (1999), Kallapur and Trombley, (1999).

3.5 Problem statement

The problem statement for this study is:

Is there any significant relationship between dividend policy and the market value of share in selected Malaysian companies?

3.6 Hypothesis Development

• Hypothesis 1:

H0: There is no significant statistical relationship between cash dividend and the market share price of selected companies in Malaysia.

H1: There is a significant statistical relationship between cash dividend and the market share price of selected companies in Malaysia.

• Hypothesis 2:

H0: There is no significant statistical relationship between earnings per share (EPS) and the market share price of selected companies in Malaysia.

H1: There is a significant statistical relationship between earnings per share (EPS) and the market share price of selected companies in Malaysia.

• Hypothesis 3:

H0: There is no significant statistical relationship between firm size and the market share price of selected companies in Malaysia.

H1: There is a significant statistical relationship between firm size and the market share price of selected companies in Malaysia.

• Hypothesis 4:

H0: There is no significant statistical relationship between debt ratio and the market share price of selected companies in Malaysia.

H1: There is a significant statistical relationship between debt ratio and the market share price of selected companies in Malaysia.

3.7 Research Methodology

The main objective for this study is to examine whether there is a significant relationship between the selected four factors and the market value of share of a corporation in Malaysia. The dependent variable for this research is the market value of share of Malaysian corporations. The independent variables for this research are the dividend policy (cash dividend) whereas the control variables are earnings per share (EPS), firm size and debt ratio.

3.7.1 Research Design

In this study, quantitative research is applied which using quantifiable and numeric data. The aim of this study is to identify the association between the independent variables such as cash dividend payment, earnings per share, firm size and debts and the dependent variable (market value of share). The deductive approach is used to formulate hypotheses and measure them empirically. According to Eldabi (2002), quantitative approach uses numerical evidence to form conclusions or to test hypothesis. Besides that, Nau (1995) stated that quantitative research design is always used to identify the truth value of propositions and enable flexibility in data treatment. For instance, statistical and comparative analysis, data collect and repeatability can define the reliability of the data. Analytical designs are used in this research for the analysis of the relationship between dividend payments and market share price of Malaysian companies to achieve the objectives of this research. Explanatory method is also applied in this research to explain the relationship between the four chosen factors and market value of share of those selected Malaysian companies. Explanatory method is used as it could recognize and meet the purpose of explaining the relationship between the independent variables and the dependent variable.

3.7.2 Research Method

The method of literature search is applied by gathering information from relevant trade publications, newspapers, magazines, annual reports and online data bases.

3.7.3 Data Source

This research uses secondary data. It is collected necessarily as it is used to examine the relationship among the variables and also applied as evidences to support the findings. The secondary data in this study refers to relevant trade publications, newspapers, magazines, corporation annual reports journals, relevant articles, Google Scholar and online data bases (EBSCO). The sources enable the collection of the market price of share of those selected companies. The published accounts of those chosen companies can be obtained from Bursa Malaysia website. Furthermore, Statistical Package for Social Sciences (SPSS) program is used to analyze and test the hypotheses by calculating the means, standard deviations, correlation coefficient.

3.7.4 Sampling Measuring Scale

The scale applied to examine the relationship among those variables is interval scale. This is because it is assumed to have equidistant points between each of the scale elements. Interval scales are also scales which are defined by metrics such as logarithms. Examples of parametric statistical techniques used in this study are Pearson correlation-r and significant (2-tailed).

3.7.5 Data Collection

Several stages are carried out in the data collection process. The first stage is to identify those selected corporations for which data are available on the announcement of the dividend date. The number of selected corporations is 103. The second stage is to collect relevant data for those selected corporations. The annual data (collected through Bursa Malaysia website) of the companies listed at Bursa Malaysia for the period of 2008 to 2012 is identified.

A total of 103 listed Malaysian companies that declared the dividend payment from the year 2008 to the year 2012 are selected as the sample for this research. A period of five years (2008 to 2012) is applied in this research. The dividend payment refers to cash dividend. The requirement is that those selected companies should have at least paid cash dividend once over the past five financial years.

3.8 Research Model

Multiple regression model and Pearson correlation analysis are used in this study in order to analyse the relationship between independent variables and dependent variable. (Baskin, 1989)

3.8.1 Multiple Regression Model

This model which primarily connects the stock price to dividend payment has been developed by other variables suggested by Baskin (1989). In this study, these variables consist of earnings per share, firm size and debt ratio.

The dependent variable (share price) is regressed on the main independent variable (cash dividend). Therefore, the equation of the multiple regressions model is given as below:

Pt = b0 + b1Dt + et

Where,

Pt = Share price for firm t

Dt = Dividend per share for firm t

et = Error term

Furthermore, earnings per share, firm size and debt ratio may also affect the dividend policy and the share price (Baskin, 1989). The stock price of small firms is usually more volatile than large firms since small firms may have less diversification and information available to investors. Moreover, leverage could have effects on the stock price due to operating risks. Also, the existence of asymmetric information links the borrowing policy to the dividend policy. As a result, the equation of the regression model is expanded as below:

Pt = b0 + b1Dt + b2EPSt + b3SIZEt + b4DEBTt + et

Where,

Pt = Share price for firm t

Dt = Dividend per share for firm t

EPSt = Earnings per share for firm t

SIZEt = Total asset for firm t

DEBTt = Debt Ratio for firm t

et = Error term

3.8.2 Pearson Correlation Analysis

Pearson Correlation Analysis is applied in this study for the analysis of relationship between independent variables and the dependent variable. This model explains the direction and relative strength of a linear relationship between two numerical variables (Harry and Steven, 1994). The values of the coefficient of correlation range from -1 for a perfect negative correlation to +1 for a perfect positive correlation.

In this study, the two variables are dividend per share distributed by those selected companies during the study period and their share prices. The calculation of the coefficient of correlation is done for each of 150 selected firms and the average calculated from the individual result. The formula is presented as below:

Where

rxy = Correlation of x and y

n = Number of items or measurement

x = First measurement

y = Second measurement

3.9 Limitation of the research

Several limitations were occurred in the process of completing this research. Those limitations may influence the accuracy of the result of findings. The major limits of this research are the size of the sample and the period of time of the study used. Most research strives for the largest possible proportion of its targeted community and for the longest period of time possible in order to get accurate and reliable results that can be generalized confidently. Excessive volatilities were occurred in the Bursa Malaysia during the year 2007 to year 2008 in term of data volatilities. The Malaysian share market was affected by the US subprime crisis and led to a rapid sell-off during the period. It will lead to a decrease in the share price due to excessive supply. Besides that, the stock market was also affected by the instability of political issues during that period. The 12th National General Election caused the FTSE Bursa Malaysia to decrease by nearly 130 points in year 2008. As a result, it is believed that the findings of this research may be affected by these stated volatilities due to the increase of variances in stock prices. Moreover, the inadequacy of research skills, expertise and limited research tools used may also influence the reliability of the findings.