Dividend Policy and payout practices in Malaysia: A qualitative analysis

Since stock prices reflect the firm’s future earnings potentials (Miller and Rock, 1985), dividends announcements therefore convey new information to the market about the future prospects of the corporation. As such, the objective of the current study is t potential role that dividend payouts play in influencing the fund managers and investors in recommending or selecting a stock, and for various stocks’ performance assessment. In addition, the study attempts to examine the possible effect of taxation on dividends payout. The study uses qualitative methods in form of semi structured interviews conducted with six Malaysian investment managers. The findings revealed that dividend payouts are not solely used as a basis for stock recommendation and assessment of companies’ performance by fund managers in Malaysia. Furthermore, taxation was found to be significant in determining dividend payouts by companies in Malaysia. These findings have great contributions to the dividend policy theory, as well as to the practitioners and policy makers that are discussed in details at the end of the paper. DOI: 10.32602/jafas.2019.


Introduction
Aspects of dividend policy relationship to agency cost models and other corporate objectives have attracted enormous discourse in academic journals and other publications in recent times. This is primarily because the regulation of agency conflicts (Jensen and Meckling, 1976) is usually impaired with various issues pertaining to a firm's dividend policies. However, the study of dividend policy relationships to other corporate events became highly important after the pioneering article by Modigliani and miller (1958) where they argued that under certain conditions, the value of firms is dependent upon its dividend policies.
Dividend decisions have also been regarded as one of the most important decisions that face modern corporations (Mollah, Keasey and Short, 2000). Agency theory precisely looks at the conflicts of interest in the principal-agent relationship. It relates the responsibilities of the managers who acts as an agent and therefore wishes to bond his activities to those of the principal who may wish in many circumstances to monitor and control the agent 1 .
Different strands of research in this area have been focused on different issues pertaining to agency relationships and other corporate objectives. Broadly, the various literatures have been focused on how to address agency problems between corporate insiders and outside shareholders of the same corporation (Easterbrook, 1984;Jensen, 1986;Myers, 1998) in relation to the corporations' dividend strategy. Agency theory suggests that corporate managers are agents of the shareholders of the corporation and their relationship is usually fraught with conflicting interest (Jensen, 1986), relating to the amount of dividends required by such shareholders. This is primarily because, such payouts to shareholders drastically reduces the amount of cash flow that will be available to managers for investments in projects that may have a zero or positive net cash flow discounted at the required cost of capital.
To reduce these agency conflicts, firms sometimes engage in different policies that include, firm's optimum monitoring and bonding package (Rozeff, 1982) and financial policy tradeoff (Crutchley and Hansen, 1989). Recently, discussions have been geared towards understanding dividend policies relationship to corporate objectives of share price maximisation.
Foremost researchers did recognize the importance dividend pay-out plays to a firm. They have suggested arguments like the fact that, it projects firm's long term stability, growth prospects, signalling future liquidity, influences the overall capital structure both at present and in the future, determines future stock price growth, etc. However, recent research has seemingly been concentrated on narrower aspects of these issues. Suggestions pertaining to its effects on the share price, its functions relating to the availability of positive NPV projects (The residual theory), etc.
The financial contributions for corporations are usually based on internal and external sources of finance. This is either through debt or equity capital that constitutes external sources of financing for corporations or by retaining the earnings of the corporation (DeAngelo, DeAngelo and Stulz, 2004). Managerial successes are mostly evaluated and tied on the amount of profits that can be generated during their entire tenure, which forms the basis for the evaluation of firm's shareholder wealth. In the circumstances, dividend payments may represent a significant aspect in evaluating managerial expertise by outsiders as it is usually tied on the level of profits that has been generated by the firm.
This research reviews the various theories relating dividend policies to various corporate objectives. Broadly, it looks at dividend policies impact on agency conflicts and share price maximisation. It also evaluates other theories that include the residual dividend theory.
The research investigates the significance of dividend policies in fund managers investment decisions, as well as the role it plays in forecasting possible changes in stock prices.
Furthermore, the study attempts to identify the possible effect of taxation and other factors on the dividend pay-out decisions. Specifically, the research attempts to answer the following questions: 1. Does dividend payment have implication to the share prices of the company? 2. What signalling effect does the payment of dividends play in various corporations?
3. Do tax payments influence companies' decisions of dividend pay-out?

The relevance/Irrelevance Hypothesis of Dividend Policies
Several theories and opinions have been either supportive or critical on the rationale being the payment of dividends by various corporations. Economists have long found dividends to be mysterious (Easterbrook, 1984). Dividends irrelevance theories argue that, dividends are irrelevant while dividends relevance theories do argue that they are relevant for investors.

Dividend Irrelevance Theory of Miller and Modigliani
This theory recognized the average investor as being indifferent with the payment of dividends and capital gains. The theory argues that under certain firm's conditions, its value is usually independent of the firm's debt to equity ratio. The theory usually termed the cost of capital theory further states that the overall cost of capital remains constant as the financial gearing of a firm increases. However, they recognized that other factors may affect a firm's dividend policy that includes, the firm's personal taxes, transaction costs, clientele effect, and so on. The specific problem here is that dividends are recognized as costly and yet ubiquitous (Easterbrook, 1984).
There is therefore no rationale in the payment of dividends by corporations and the harder corporations try to pay dividends, the more puzzling it becomes. The irrelevance theory argues that, firms that reduced dividends payout prosper (better off) relative to others (Easterbrook, 1984). Following this pioneering research by Modigliani and Miller (1958), researchers have argued that the value of a firm is dependent on its dividend policies.
However, recent researchers have argued that the value of a firm does not rely on its dividends policies but on the potential earnings power of the corporation.

Dividends and the Firm's Capital Structures
Researchers have long argued on different aspects of the ability of dividend policies in determining firm's capital structures. Following Faulkender and Petersen (2005), the two major financing policies of various corporations are the relationship relating its dividend policies to its capital structures. Suggestions purported by Jensen (1986) argues that dividends can be used as a way of retiring a firms' debt burden rather than wasting available and free cash flow on unprofitable investments. Additionally, research has treated dividend policies and aspects of capital structure as two distinct policy variables, but there is several research that have suggested that both have common factors and that corporations treat both as part of financial planning determination (Faulkender and Petersen, 2005).
A very important part of recent theories is that dividend policies and capital structures are jointly determined by control allocations between managers and investors and variations in both are therefore driven by the same underlying factors and assumptions (Faulkender and Petersen, 2005). A very important issue that shapes dividend policies is the conflict of interest between bondholders and shareholders. While shareholders will want to be paid more dividends, bondholders will always think that they are trying to expropriate wealth from them. However, it has been suggested that this conflict can be mitigated when firms hold more collateralizable assets because these assets may serve as collateral against borrowing (Titman and Wassels, 1988).

Dividends and the Residual Dividend Theory of the Firm
This model argues that firms usually base their dividend policies on their cash flows needs and capital structure decisions (Miller and Modigliani, 1961). For example, when they recognized positive NPV projects, they will prefer to retain some of the corporation's earnings for the investment in such projects and the residual cash flow will be utilized to pay dividends. On the other hand, when they do not recognize positive NPV projects, they may decide to pay dividend when free cash flow is available.

Dividend Policy Relationship to Agency Theory
Researchers have recognised different factors, as having a significant impact on a firm's dividend policy. Following the seminal study of Jensen and Meckling (1976), other studies have provided realistic arguments linking agency cost as playing a significant role in the determination of firm's capital structure, other financial activities and its agency cost. Easterbrook (1984) suggested that the payment of dividends by firms reduces corporate agency cost. Following the agency model developed by Allen, Bernardo and Welch (2000), large shareholders can use information they discover about their firm's qualities to reduce the potential agency problems.
Furthermore, Jensen (1986) recognized that the payment of dividends is better for institutions as the distribution of free cash flow to stockholders and other shareholders is usually better than investments in negative and unprofitable net present value projects.
Researchers (Jensen and Meckling, 1976;Rozeff, 1982;Easterbrook, 1984;Crutchley and Hansen, 1989;Jensen, Solberg and Zorn, 1992) have long suggested that the payment of dividends have a significant impact on the agency problem 2 between manager and shareholder by reducing the discretionary funds available to managers. Several years ago, research by Shleifer and Vishney (1986), recognised that dividends can be used as a mechanism to compensate company's institutional investors. Another kind of conflict exists between shareholders and corporate bondholders as it has been suggested that shareholders can voluntarily appropriate bondholder's wealth through a voluntary payment of high dividends to themselves (Smith and Warner, 1979).

Conflict of Interest Connecting Dividend Payments to Corporate Objectives
Following Jensen (1986), corporate managers sometimes have divergent objectives to those of shareholders and other stakeholders. In this setting, they will presumably not be able to employ the utilization of incentive contracts to enhance their managerial opportunism but paying out cash in the form of dividends may be a reliable way of expressing such opportunism.
The more a firm is financed by outside or debt capital, the more it is subject to more capital market and other institutional controls (DeAngelo et al., 2004) thereby leading to potential agency problems involving disputes between the shareholders, who are outside contributors to the firm's capital and the firm's management. In these circumstances, firms will prefer to finance future project through retained earnings rather than through external capital. Several suggestions have been of the opinion that dividends should be used to compensate institutional investors since institutional shareholders do have a probability to find value improvement investments (Allen et al., 2000).
Managers frequently have the choice between satisfying the aspirations of institutional shareholders and share price maximisation. When firms do pay higher dividends, they attract higher institutional ownership, and such institutions do in turn require playing a larger role in overseeing management than retail investors. Managers will therefore have the choice to weigh the effect of the positive relationship to the share price response when dividends are announced against annoying institutional investors through a reduction in dividends as responses to poorer performance. Additionally, the probabilities of paying dividends increase with the amount of equity earned in the capital structure of the corporation (DeAngelo et al., 2004).

Empirical evidence
Researches relating to the reasons why firms pay dividends have been pervasive. Despite the unfavourable tax treatment of dividends, researchers have long recognised that firms and shareholders are frequently focused on the amount of dividends to be paid by various corporations. A wholly accepted view suggests that dividends are valuable signals of firms' prospects and future profitability (Bhattacharya, 1979;Miller and Rock, 1985). Dividends can be used to attract institutional shareholders of the firm as the presence of large institutional shareholders usually signal firm quality and are properly managed (Allen et al., 2000). Dividends may also be used to control insiders thereby limiting their frequent lack of control (Easterbrook, 1984;Jensen, 1986) in for example, managing free cash flow.
Otherwise, dividends may be preferred because they may influence a lower transaction cost relative to the firm's capital gains since investors sometimes view income from dividends different from those capital gains (Perez-Gonzalez, 2003).
Dividends are very important for companies and company boards usually declare regular dividends and raise them from one period to the other. Many managers recognise that a higher dividend leads to higher share prices for their companies (Easterbrook, 1984).
However, progressive firms may be more willing to withhold dividends and use cash generated for reinvestment because internal financing is cheaper than issuing dividends and floating new securities (Easterbrook, 1984). Similar dividend policy aspects including the perceived relationship between dividend payment policy and share values, the impact of taxation and attitudes to share buybacks were also explored by McCluskey, Broderick, Boyle, Burton and Power (2010) in the Irish context. The authors found that dividends are an important in investor decision-making processes and that dividends influence share valuations. Another key finding is that fund managers appear to be able to influence the dividend policy of Irish companies in which they have a shareholding. Finally, taxation issues appear relatively unimportant and the majority of fund managers prefer cash dividends to buybacks.
In the Malaysian context, Lee, Isa and Lim (2012) investigated the relationship between dividend changes and future profitability of firms. The authors found that dividend changes are strongly related with contemporaneous earnings changes, weakly related with one year ahead of earnings changes and largely unrelated with earnings changes beyond one year.
Further, the authors found weak evidence that the size of dividend changes is related to future profitability.
In the same context, Zainudin, Mahdzan and Yet (2018) analysed the relationship between stock price volatility and dividend policy of industrial products firms listed on Bursa Malaysia. The study covered a sample of 166 public-listed firms spanning from 2003 to 2012, and used Baskin's framework (Baskin (1989). The findings revealed that earning volatility significantly explains stock price volatility during the crisis period, while dividend payout ratio predominantly influences volatility during pre-and post-crisis sub-periods.
These studies shed light on dividend policy and behaviour among firms listed in the Malaysian stock market. Nevertheless, further analysis is required to comprehensively examine the dividend policy practices by Malaysian fund managers.

Methodology
In line with the abovementioned objectives, a qualitative research approach is used.
According to Merriam (2009), qualitative research allows the researcher to understand how people interpret their experiences, how they construct their worlds, and what meaning they attribute to their experiences (p.5).
The choice of qualitative research methodology can be further explained by its ability to generate comprehensive details about dividend policies and payouts and how they are perceived and used by various fund managers and investors. In contrast to brief answers to structured questions which will not be able to provide the required in depth information to adequately assess the issue at hand (Weischedel, Matear and Deans, 2005).
As such, an initial list of interviewees was established covering some of the investment companies in Malaysia that are perceived to be suitable for the study. A total of six interviewees were subsequently selected based on specific criteria, namely, experience in the fund management and investment field, exposure to the main issues of dividends payouts and policies, and educational level. Particularly, the selected interviewees are at least holders of a Master's degree in related specialisations (mainly finance and economics). Moreover, the interviewees have a minimum of five years of experience in investment management. This allows the interviewees to understand and respond to detailed questions concerning the issues related to dividend payouts. In this regard, Polit, included in the study, to allow an in-depth exploration of topics in phenomenological studies. Furthermore, the sample of six respondents is considered suitable since a similar sample has been used in comparable studies (Tijani, Fifield and Power, 2009;Koenigstorfer and Klein, 2010). It is worth noting that all the interviews were reviewed several times before been transcribed. Subsequently, a phenomenological approach to analyse data was adopted, which involves interpreting and reflecting on the data transcript so as to achieve a holistic understanding of the meaning of the participants' experiences (Alexis and Vydelingum, 2007).

Summary of findings
This section presents a summary of the main interview findings on three successive but related themes, namely, the extent of importance of dividend policies for shares' expected performance and selection by fund managers. Secondly, the signalling effects of dividend payouts and their implications, and finally, the role of taxation in determining dividend payout and policies.
Firstly, it goes without saying that information is the most important and precious commodity in financial markets, and one of the most important information an investor can obtain in financial markets are earnings and dividend announcements. A considerable number of studies have demonstrated that indeed dividend announcements have a significant effect on securities' selection (Miller and rock, 1985;McCluskey et al., 2010). In line with this argument, most of the interviewees were of the opinion that dividend announcements can sometimes be significant in understanding the future trend of a company's stock; however it cannot be used as a basis to make investment recommendations. Indeed the interviewees noted that in many cases, the announcement comes in contradiction with the results of the financial analysis performed by the investment company and the recommendation to be made. This can be simply explained by the fact that sometimes a dividend payout is a strategic financial decision made by the company to boost investors' confidence in its stock, in an attempt to manipulate the market's perception about the company's stock, regardless of its actual performance, and away from the profits actually realised in a given year (Aroni, Namusonge and Sakwa, 2014;Olang and Grace, 2017). Hence, announcing a dividend payout by companies has had varying implications in the case of Malaysia. For instance, interviewee E noted that dividend payouts are usually included as one of the parameters for recommending a stock when there is a sustainable and growing trend of dividend payments, and when the dividend announcement is in line with the rest of the considered parameters.
On the other hand, dividend payout has been seen as a powerful signalling factor under the signalling hypothesis, and this has been demonstrated in several empirical studies (Brickley, 1983;Czapiewski and Kubiak, 2018). However, it was noticed through the interviews that most of the interviewees disagreed with this principle. To most of them there is more significant reliance on free cash flows to gain insight on the future trend of stock. And evidently dividend or earnings announcements are not always equivalent to positive performance by the company. In other words, they do not have a direct effect on the stock price. Hence, most of the interviewees disagree with the dividend signalling hypothesis in the Malaysia context.
Finally, dividend payouts have always been associated with the taxation systems and tax payments. Specifically, the level of tax might determine the amount of dividend that the company would distribute to shareholders, and hence the net dividends to be actually received. At this level, most of the interviewees perceive that taxation is practically significant in influencing the dividend payouts by companies. Hence, this has the potential of directly causing agency issues, as the payment of dividends is suggested to eventually reduce the agency issues and costs (Easterbrook, 1984;Ghosh and Sun, 2014). This finding contradicts the findings by McCluskey et al. (2010) who found that taxation issues are relatively unimportant for the majority of the Irish fund managers. This might be explained by the difference between the Malaysian and Irish financial markets and economic conditions, which will eventually have an impact on the investment style and determining criteria.

Discussions and conclusion
The main objective of the study was to explore the importance of dividend payouts in investors' selection and recommendation of companies' stocks as well as the eventual signal that dividend payouts might provide to investors on the future trend of stocks.
Finally, the study attempted to examine the possible effect of taxation on dividend payouts.
To achieve these objectives, in-depth semi-structured interviews were conducted with six investment managers in Malaysia. Overall, the findings revealed that dividend payouts are not solely used as a basis for stock recommendation and assessment of companies' performance by fund managers in Malaysia. Furthermore, taxation was found to be a significant in determining dividend payouts by most companies in Malaysia.
These findings have significant implications for the dividend theory, as well as for the practitioners and policy makers. Firstly, it shows that the dividend signalling hypothesis is not valid in all settings. It largely depends on the nature of the companies and their respective sectors, but it mostly depends on the trend of dividends payout and their historical magnitude and sustainability. This finding further supports the dividend irrelevance theory at least in the Malaysian context, as most of the interviewees usually disregard the dividend payouts in recommending and assessing stocks' performance.
On the other hand, the policy makers should take these findings into account to enhance financial investment, especially regarding taxes. It would be very significant to develop a comprehensive tax system that fairly and accurately accounts for dividends received from different countries for the benefit of the shareholders as well as the original country's authorities. In addition to that, fund managers as well as independent investors in Malaysia should be aware that dividends' announcements do not always signal the proper indication they are expected to, sometimes they intend to send a reversed message. Hence, if they consider dividend payouts as one of the factors for selecting a stock, then they need to always have a comprehensive analysis that include a set of other factors as well. Moreover, they have to watch these dividends for a long period time, long enough to establish a trend of development in payouts that might give them a hint on the possible future developments of different companies' shares.
Though the current study has significant contributions, there also some limitations that have to be noted. Firstly, the study employed qualitative analysis only in the form of interviews and document analysis. Certainly, the study could have amounted to more significant findings if both qualitative and quantitative methods were used and triangulated. Hence, this is highly recommended for future studies. Secondly, the study focused only on the Malaysian context which renders the findings ungeneralizable to other settings. As such, future studies are highly recommended to extend these findings to more diverse settings for generalization purposes. Finally, the study focused only on taxation as a factor that influences dividend payout, while the other possible factors were not emphasized in this study. Thus, the future studies are highly recommended to focus on the other factors that can possibly affect the dividend payouts in various regions.