DETERMINANTS OF DIVIDEND PAYOUT RATIO: EVIDENCE FROM LISTED COMPANIES ON HO CHI MINH STOCK MARKET In Partial Fulfillment of the Requirements of the Degree of MASTER OF BUSINESS ADMINISTRATION By Mr. Hoang Ba Nghi ID: MBA05025 International University - Vietnam National University HCMC August 2014 DETERMINANTS OF DIVIDEND PAYOUT RATIOS: EVIDENCE FROM LISTED COMPANIES ON HO CHI MINH STOCK MARKET In Partial Fulfillment of the Requirements of the Degree of MASTER OF BUSINESS ADMINISTRATION By Mr. Hoang Ba Nghi ID: MBA05025 International University - Vietnam National University HCMC August 2014 Under the guidance and approval of the committee, and approved by all its members, this proposal thesis has been accepted in partial fulfillment of the requirements for the degree. Approved: ---------------------------------------------Chairperson ---------------------------------------Committee member ---------------------------------------------Committee member ---------------------------------------Committee member ---------------------------------------------Committee member ---------------------------------------Committee member i Acknowledge This thesis report is about the Determinants of dividend payout ratios: Evidence from listed companies on Ho Chi Minh Stock Market for the period of 2008-2012 would not have been possible without valuable contribution of all teachers from school of Business. I would like to thank to the International University, Vietnam National University – Ho Chi Minh City for giving us a great opportunity to practice and learn more knowledge, especially in the field of finance. I especially appreciated the School of Business that helped me have good condition to do the necessary research work and archive the results. I am deeply indebted to my thesis project advisor Dr. Nguyen Van Phuong, for his patience, guidance and advice throughout this plan. He was always keeping his eyes on my research. This gave me the efforts which proved valuable for the success of this thesis project. Finally, I would like to thank my parents, also my friends for supporting and encouraging me throughout my studies. With their love, I could finish this work. I hope this will serve as a valuable resource for whose major or carrier related to this field. ii Plagiarism Statements I would like to declare that, apart from the acknowledged references, this thesis either does not use language, ideas, or other original material from anyone; or has not been previously submitted to any other educational and research programs or institutions. I fully understand that any writings in this thesis contradicted to the above statement will automatically lead to the rejection from the MBA program at the International University – Vietnam National University Ho Chi Minh City. iii Copyright Statement This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognize that its copyright rests with its author and that no quotation from the thesis and no information derived from it may be published without the author’s prior consent. © Hoang Ba Nghi / MBA05025 / 2014 iv Table of Contents Chapter One - Introduction .......................................................................... 1 1.1 General ideas ...................................................................................... 1 1.2 Rationale ............................................................................................. 3 1.3 Problem definition .............................................................................. 4 1.4 Objectives ........................................................................................... 5 1.5 Research questions ............................................................................. 6 1.6 Contributions ...................................................................................... 6 1.7 Scope and limitations ......................................................................... 7 1.8 Recommendation for future research ................................................. 8 1.9 Structure of the paper ......................................................................... 8 Chapter Two - Literature review ................................................................. 9 2.1 Definitions .......................................................................................... 9 2.2 Theories of dividend payment ............................................................ 10 2.2.1 Modigliani-Miller Dividend Irrelevance Theory ..................... 10 2.2.2 The “Bird in Hand” Theory ..................................................... 13 2.2.3 Signaling Theory...................................................................... 15 v 2.2.4 Agency Theory......................................................................... 17 2.3 Determinants of dividend payout ratio and hypothesis ...................... 19 2.3.1 Financial leverage .................................................................... 19 2.3.2 Investment ................................................................................ 20 2.3.3 Risk .......................................................................................... 20 2.3.4 Corporate income tax ............................................................... 21 2.3.5 Liquidity................................................................................... 22 2.3.6 Firm size................................................................................... 23 2.3.7 Structure of asset ...................................................................... 23 2.3.8 Sales Growth ............................................................................ 24 2.3.9 Cash flow ................................................................................. 25 2.3.10 Market – to – book value ....................................................... 25 2.3.11 Ownership concentration ....................................................... 26 2.3.12 Return on equity..................................................................... 27 2.4 Model for research.............................................................................. 29 2.5 Hypotheses ......................................................................................... 30 Chapter Three – Research Methodology ..................................................... 32 vi 3.1 Problem statement ............................................................................. 32 3.2 Research objectives ........................................................................... 33 3.3 Data ................................................................................................... 33 3.4 Research method ............................................................................... 35 3.5 Data analysis procedures .................................................................... 36 3.6 Data analysis types and methods ...................................................... 36 3.7 Sample collection and data analysis ................................................. 38 Chapter Four: Data analysis ......................................................................... 39 4.1. Sample description ............................................................................ 39 4.2. Panel data models ............................................................................. 41 4.2.1 Fixed effect – Hausman test ..................................................... 41 4.2.2 Checking correlation ................................................................ 43 4.2.3 Pooled OLS analysis ................................................................ 46 4.2.4 Testing multi-collinearity......................................................... 47 4.2.5 Linear regression with large dummy variables set .................. 49 4.2.6 Linear regression with lagged dependent variable................... 51 4.3 Discussion .......................................................................................... 54 vii 4.4 Answer research question ................................................................... 58 4.4.1 Question 1 ................................................................................ 58 4.4.1.1 Financial leverage ................................................................. 59 4.4.1.2 Return on equity.................................................................... 60 4.4.1.3 Corporate income tax ............................................................ 61 4.4.2 Question 2 ................................................................................ 61 Chapter Five – Conclusion ............................................................................ 64 5.1 Conclusion .......................................................................................... 64 5.2 Contribution........................................................................................ 65 5.3 Recommendation ................................................................................ 66 References .................................................................................................... 68 viii List of table Table 4.1 Descriptive Statistic from 2008 to 2012 .......................................... 40 Table 4.2 Hausman test result ......................................................................... 43 Table 4.3 Correlation matrix ........................................................................... 45 Table 4.4 OLS regression result ...................................................................... 47 Table 4.5 VIF result ......................................................................................... 48 Table 4.6 Linear regression with large dummy variables set result ............... 50 Table 4.7 Linear regression with lagged dependent variables ........................ 53 Table 4.8 Result comparison............................................................................ 55 Table 4.9 Comparison between empirical evidence and results...................... 56 ix Abstract The main purpose of this paper research is to find out the association between dividend policy and financial leverage, investment, risk, corporate tax, liquidity, size, structure of asset, sale growth, cash flow, market-to-book value, ownership concentration, sale size, and return on equity of listed companies on Ho Chi Minh City Stock Exchange (HOSE). The sample consisted of 301 listed firms on HOSE in five years range, from the year of 2008 and 2012. The study is carried out in form of quantitative research approach with a deductive method. And we have based on some previous dividend theories including: the dividend irrelevance theory, the bird in hand theory, the signaling theory and the agency theory. So as to examine whether it exists the relationship between the selected factors and the dividend payout ratio, the paper uses both an Ordinary least square (OLS) and a linear regression. Multi-collinearity tests were also analyzed so as to prove that no multi-collinearity affected the results of the study research. The result indicates that some of the selected determinants have an impact on the companies’ dividend payout ratios. Especially, return on equity is the strongest positive impact on dividend payout ratio. Keywords: Dividend policy, Financial leverage, Investment, Risk, Corporate tax, Liquidity, Size, Structure of asset, Sale growth, Cash flow, Market-to-book value, Ownership concentration, Sale size, and Return on equity. x This page is intentionally left blank 1 Chapter One - Introduction The main purpose of this first chapter is to provide a framework for this research paper with details following in subsequent chapters. A major component of the chapter is a statement of the research problem to be examined in the following chapters. 1.1 General ideas In corporate finance, dividend policy is paid much attention on. Based on the business perspective, the dividend is a significant source of money to be settled outside. For instance, in particular 1995, the listed companies on the New York Stock Exchange had paid up the amount up to USD 130 billion of dividends for shareholders. In terms of research, the dividend policy is a matter of controversy. The question of whether businesses should pay dividends to shareholders or retain that income to reinvest to other projects for earning more profit purpose remains unclear. According to Lease (2000, p.29), the term of ‘dividend policy’ refers to the practice that management follows in making dividend payout decisions. Normally, dividend is defined as a sum of money from the company’s profits paid commonly to investors. In another way, the dividend is the profit-sharing to the company’s investors, compensating them back part of the capital. Likewise the strategic options, the dividend is considered as an easy way for investors to get profit from possessing stocks without trading them. Normally, the board of directors gives the decision of how much should pay as dividend to the shareholders of a firm. The matter of dividend payout ratio continues as one of the most challenging and dubious issues in corporate finance and financial economies. The company can reinvest in 2 operations in future with the hope of creating more earnings, and market price of the company in the future will be more appreciated. Either the company will payout a certain percentage of profits to stockholders under the form of dividends. In the case of the first option, the company should maintain consistent and instant growth to demonstrate a convincing manner that the re-investment will generate a rate of return for investors greater than receiving dividend. In general, the company’s slow development means that the financing opportunities in future are not appreciative, its stock price will not be increased much more and the company should pay dividends to shareholders. The dividend enigma has been an enduring problem in finance without being resolved. Allen, Antonio and Ivo (2000) outlines the current harmony view when they come to the conclusion “Although a number of theories have been put forward in the literature to explain their popular presence, dividends remain one of the most enigmatic matters in corporate finance”. A few theoretical literatures on dividend payout policy have conciliated the unquestionable logic of their dividend irrelevance axiom with the assumption that both managers and markets recognize payouts either in stock or cash as form of dividend. The academic work on this issue proposed five possible weaknesses that management should consider, like as taxes, asymmetric information, incomplete contracts, institutional constraints, and transaction costs (Allen et al., 2000). Previous observational literatures have found out several determinants that are critical for dividend payout policy. Lintner (1956) describes the dividend payment arrangement of a corporation is substantially attributed to current year earnings (E0) and preceding year dividends (Dt-1). 3 Pursuant by Modigliani and Miller (1963) theorem, the reaction of tax on dividend payout policy has been examined a lot of literatures. Nonetheless, primary experimental investigations have been checked through linked variables that are necessary to describe and conclude what factors have remarkable impact on dividend payout policy (Black and Scholes, 1974; Miller and Rock, 1985). In general, developing firm holds retain earnings for its internal funding and pay fewer dividends since in this concern the return on investment is high. Consequently, it becomes an imperative issue of interest in financial literatures to investigate how much to pay to the shareholders in cash or in stock. The former, that is, the cash dividends are two ways to distribute cash to stockholders: stock repurchases or dividends. Such dividends are paid out in cash and are usually taxable to the beneficiary in the year they are paid. The latter, that is, the stock dividends are those made payment in the form of supplementary stock/stocks of the issuing firm. Commonly, stock dividends are issued in fraction to share owners. To support and to have ideal measure of the factors analyzing dividend payout policy, the study consults both cash and stock dividend. Several intellectual publications have pursuit to build up theoretical models that managers should consider when making dividend payout decisions. Only a little literature, especially for dividend determinants in Vietnam has focused on empirical estimations. 1.2. Rationale It is generally recognized that the particular choice of dividend payout by a business at a given point of time has important implications for a variety of considerations of significance to management, particularly those of future investments and attraction of new equity capital. 4 Dividend, because of its nature of cash disbursement, is normally thought as contributing to the opportunity for getting attention of current and potential investors. Dividend policy is an important part of the financial policy of the enterprise as it has not a direct impact on the benefits of the shareholders, but also influence on the company’s development. Around the world, there are many different studies on dividend policy, but it is still a problem even more controversial. Along with the development of the Vietnam stock market, dividend policies of Vietnamese enterprises increasingly play an important role. They have recently emerged as one of the issues being much focused by investors, and businesses have not yet been acutely highlighted the emphasis of the dividend policies of its financial activities, therefore it has revealed some problems and shortcomings. This research hopes to get the real picture being painted dividend policy of state enterprises in Ho Chi Minh City stock market in the past five years and thereby points out the problem existing, as well as issues rational proposals for market. Wishing to contribute for research importance elements to the goal of maximizing shareholder assets , as well as further analysis of dividend distribution decisions, I would like to choose the topic: " Determinants of Dividend Payout Ratios: Evidence from listed companies on HOSE "to study science subjects. 1.3. Problem definition Companies can use internal or external sources to manage their investments. Internal sources comprise retained earnings and depreciation, while external sources radically indicate new borrowings or stock issue. Hence, the funding decision comprises the evaluation of two choices. The first choice is the dividend choice (the ratio of retained earnings to be reinvested and the ratio to be paid out as dividends). 5 The other is the capital structure choice – the portion of external finance to be obtained and the portion to be raised in the form of new equity. There is an imperative gap in modern finance theory on the issue of corporate dividend policy. The doctrine should be able to explain the reason why dividends signal both “good” and “bad” news. For years, the term "information content of dividends" has been frequently used in finance literature (Watts, 1973; Laub, 1976 and Venkatesh, 1989). The phrase refers to the hypothesis which shows that dividends transfer information about future interests - information that allows market participants to predict future earnings more precisely. Lintner (1956) is one of the first to suggest that current dividends depend on future as well as current and past earnings. Subsequently, a debate over the dependence of a firm's market value on the degree at which dividends are paid out of earnings (dividend payout rate) developed in the literature. Modigliani and Miller (1963) demonstrated, under the premises of pure capital markets, rational behavior, and duty-free, that the value of the firm does not rely on the firm’s dividend payout rate. Previous studies being conducted on dividend payout do not agree that whether payouts signal positive or negative conditions about a corporation (Watts, 1973). Also, prior empirical evidence regarding the impact of dividend taxes on firm valuation is mixed (Gentry, William, Deen, and Christopher, 2003). 1.4. Objectives Thematic objective is based on the theories and practical evidences of dividend policies around the world to reach the analysis and evaluation of the dividend policy of listed companies on the HoChiMinh City stock market, which can be representative 6 for Vietnam stock market. Since then, a significance proposal is set up for building appropriate dividend policy for the listed companies in the future. Therefore, the research will mainly focus in declaring the following objectives:  To examine the factors that determine the dividend payout ratio in HOSE’s market  To analyze the impact of proposed determinants on dividend payout ratio policy. 1.5. Research questions Thus, the research questions of this study are: - What is the specific relationship between the dividend payout ratio and selected factors for listed company in HOSE? - Whether the degree of significance determinants on payout ratio policy is homologous according to collected data? 1.6. Contributions This research just adds a little bit of something to an already existing work, its main target is to make a valuable contribution by innovating on the rich existing literature. One of the proposed contributions of this paper is to investigate the explanatory variable’s signs and their significance. An important usefulness of this paper is to find out whether there are any significant differences in the parameter values over the time period 2008 to 2012. Both theoretical and empirical approaches are undertaken with the purpose to contribute a more comprehensive aspect on the subject. Particularly, literature review is devoted to a review of previous theoretical and experimental literature on the dividend decision controversy around the world. This study attempts to innovate on existing academic work is by concentrating on 7 emerging, particular HoChiMinh City market, as opposed to developed. As noted in the literature review part, less of the theoretical and observational studies that deal with financing decisions are Vietnam based. Finally, the study builds decision models of dividend policy in the business, and proposes appropriate selection dividend policy for firms listed on the stock market in HoChiMinhCity. 1.7. Scope and limitations The research is only the synthetic observations, analysis bases on data being published in the mass media without the survey questionnaire of measuring the impact of these factors to decide dividend policy of listed companies in Vietnam. This study is for the listed companies in particular; all joint stock companies in general, a proper level of awareness and rational importance of dividend policy and its impact to value businesses through theoretical analysis. Therefore, this study ignores many of the theoretical views of other dividends. For example: Brennan theory (based on the model of capital asset pricing (CAPM), Brennan announces that pre-tax income of a stock has the linear relationship with the systemic risk and the ratio dividend, and the ratio dividend, higher pre-tax income is to offset the taxable dividends that shareholders receive) or theoretical Elton and Gruber (given that the reduction in the share price on the ex dividend low than the dividend per share). This study remains obvious limitations being needed to be overcome, supplement and deeper analysis to be more complete, more highly applicable to partially contribute to the overall development of Vietnam stock market in the future. 8 1.8. Recommend for future research Based on the literature explored, the paper aims to identify the factors determine the dividend policy for companies on HoChiMinh City stock market. By studying dividend policy of these companies using elements model, it is hoped to create a basis for future study of dividend determinants of other industry in Vietnam too. The subject can be extended by the analysis of other policies by stock dividend, buying back shares and evaluating the combined effects of all of these dividend policies and corporate values. The empirical findings have significant roles for operators and policy makers to apply the same methodology to determine the determinants of dividend policy factors of other industries in Vietnam. 1.9. Structure of the paper Chapter 1: Introduction, presenting the foundation to set up the topic, problem statement, objectives, scope, significance and contribution of the paper. Chapter 2: Literature review, presenting theories, definitions related to dividend policy, exposing model and research results of previous research related to this topic. Suggesting research model for this paper. Chapter 3: Research Methodology, presenting the research approach, the way to take sample, measures, method of analyzing data. Chapter 4: Data analysis, analyse the data and provides the descriptive statistics table for all the regression variables. Chapter 5: Conclusion, summarizing the research results of the topic -the determinants of dividend payout based on companies listed on HOSE. 9 Chapter Two – Literature review The main purpose of this chapter is examining historically significant research studies that act as the basis for the proposed study. With a view to implement the main purpose of this paper, the component problem statements are identified as follow: “Determinants of dividend payout ratios: Evidence from listed companies on HCM Stock market” This literature search is carried out to find materials relevant to dividend payout ratios, and other determinants that have the relationship with such as: financial leverage, investment, market-to-book value ratio, cash flow, liquidity, ownership concentration, sales growth, firm size, return on equity, corporate tax, risk, structure of asset. These materials include books, ebooks, materials issued by advisors and so forth. 2.1. Definitions Investopedia defines “dividend payout ratio” as how well earnings approve the dividend payments. In the U.K., they often use a similar ratio called dividend cover. It is calculated as earnings per share divided by dividends per share. The dividend payout policy is really one of the most controversial subjects within corporate finance field and some scholastics treat the company’s dividend payout policy an unsolved puzzle. Although a large scale amount of research regarding dividends has been conducted, there is no homogeneous response for the question: what are the determinants of the companies’ dividend payout ratios? There exist many reasons for 10 why companies should pay or not to pay dividends. For example, the dividend payout is critical for investors because:  Dividends present certainty about the company’s financial welfare,  Dividends are attractive for investors looking for securing current income, and  Dividends help preserve market price of the share. Companies that have an enduring past record of unchanged dividend payouts would be negatively influenced by lowering or eliminating dividend distributions. These companies would be positively influenced by upgrading dividend payouts or making supplementary payouts of the same dividends. Moreover, companies without paying a dividend in the past are perceived favorably when they inform new dividends. A decline dividends payout is looked badly by investors, and the stock price usually decreases as investors pursue other dividend-paying stocks. An unfluctuating dividend payout ratio illustrates a solid payout policy by the board of directors. The dividend payout ratio supports an idea of how well earnings approve the dividend payments. More grown-up companies favor to have a higher payout ratio. 2.2. The theories related to dividend payment 2.2.1. Modigliani-Miller Dividend Irrelevance Theory Franco Modigliani and Merton Miller (1961) present one of the most influential dividend theories and even though it is generated for more than 50 years ago, it is still seen as one of the most respected theories. When the theory is presented in the article “Dividend policy, the valuation of shares and the growth” it provides a new 11 benchmark and changes the view that both practitioners and academics have towards dividends. Before the publication of Modigliani-Miller‘s dividend irrelevance theory, the general view is that dividends are highly correlated to the value of the stock (Baker, 2009, p.98). As the name of the theory suggests, it states that under perfect capital markets, the dividend policy is liberated to the value of the company, and it does not issue whether the company has high or low dividend payments. Modigliani and Miller (1961) use three criteria in order to define a perfect capital market:  Perfect capital market - no single actor on the market is immense enough to affect the market price of a guarantee security and everyone has entry to the alike costless information, i.e. no actor has an information advantage. Another important presumption is that there are no transaction costs or taxes and all actors can operate on the market under the same conditions.  Rational behavior – it only states that all actors on the market prefer more belongings to less. It also assumes that it does not matter whether the actors receive the rise in revenue in the pattern of capital gains from the dividend or stock payments.  Perfect certainty - all actors on the market has the same report and knows the return of each security in the future. Therefore, it is possible to make the assumption that there only exists security which Modigliani and Miller assign to as stocks. In respect to the assumptions discussed above, the dividend payments become irrelevant for the shareholders. Because of paying dividends, the company has to issue new shares in order to raise the needed capital. As the new stocks are issued, the price of the stocks will drop in 12 equal proportions to the dividend payments and the decrease in stock price and the dividend payments will cancel each other out. For example, if the company pays a dividend of 10 VND the shareholders receive 10 VND for each of the shares owned. Simultaneously the stock price will decrease 10 VND since more shares are issued in order to raise capital, the shareholders are therefore equally well off no matter the dividend payments. Modigliani and Miller also explain that the shareholders are capable of constructing their own homemade dividends. Like for instance, if the company does not pay dividends but the shareholder prefers 2 percent dividend he can sell 2 percent of his stocks and thus creates a homemade dividend. The adverse is of course also true, if the company pays a higher dividend than the shareholder prefers he can use the surplus dividends to buy additional stocks (Brigham & Houston, 2011, p.484). These two arguments discussed above are the underlying assumption of the irrelevance hypothesis and according to these arguments shareholders should be indifferent between capital gains and dividends. This in turn contributes to that the shareholders are unwilling to pay a greater price for dividend paying stocks which in turns make the question of dividends irrelevant. According to M&M theorem, under perfect condition of capital market, the company’s dividend payout policies do not influence the value of a firm. A number of previous research studies examine the validity of Modigliani and Miller’s theories and the propositions have both been supported and rejected by various academics. Black and Scholes (1974) favors the results of Modigliani and Miller and states that companies are able to adjust dividend payments in accordance with the preferences of tax-induced investors, and for this reason there are no relationship between dividends and stock returns. Another prominent research made by Miller, Merton and Scholes (1978) also maintain the 13 propositions in the dividend irrelevance theory and they identify that even though the tax rate for dividends and capital gains are different under the US tax system the dividends do not involve the value of the corporation. 2.2.2. The “Bird in Hand” Theory The opposing view towards Modigliani and Miller’s dividend irrelevance theory is that dividends affect the company’s value and this assumption is represented by the so called “bird in hand theory”. The theory is first mentioned by Lintner (1956) and it has been supported by various researchers including Gordon (1959&1962). Hence it is one of the most well-known and respected dividend theories. The name “bird in hand” is the umbrella term for all studies that argues that dividends are positively correlated to the company’s value. Expressed in financial terms, the theory says that investors prefer to invest in stocks that pay current dividend rather than to invest in stocks that retain earnings and pay dividends in the future. This is because of the high degree of uncertainty related to capital gains and dividends paid in the future. Current dividends are more predictable than capital gains, as the stock price is ascertain by market forces and not by the managers it has a higher degree of uncertainty (Keown, Martin, Petty & Scott 2007, p.418). Gordon’s (1962) dividend model is based on the following assumptions:  The company is all equity financed and no external financing is used. The company finances all investment with retained earnings.  The retention ratio, cost of capital, and internal rate of return is constant.  The company has an eternal life. 14 The underlying assumptions of Gordon’s model is based on the idea of what is available today compared to what may be available in the future (Khan & Jain, 2008, p.30). It is based on the logic that the distant the future is, the higher the uncertainty regarding capital gains and future dividends is. Even though the capital gains in the future may provide a higher return than the current dividends, there is no guarantee that the investor will accumulate a higher return due to the high degree of uncertainty (Gordon, 1962). Since the length of the time and the level of risk are correlated, investors are unwilling to invest in companies where the time until the dividend payments are far away. Therefore an investor would be willing to pay a higher price for firms that pay current dividends. For companies who do not pay current dividends, the investor would use a higher discount rate in order to discount the earnings and the value of these companies should therefore be lower than the companies who pay current dividends (Khan & Jain, 2008, p.30). The opposite is of course also true, companies who pay current dividends have a lower level of retained earnings which contributes to lower discount rate which in turn contributes to a higher value of the firm. The bird in hand theory has been subject to a large amount of criticism and opponents to the theory states that it excludes important factors. Keown et.al (2007) argues against the theory and says that increases in current dividends do not decrease the riskiness of the company; it does in fact work in the opposite direction. Because if an increase in dividend payments are made, the managers have to issue new stocks in order to raise the needed capital. Therefore a dividend payment just transfers the risk from the old to the new shareholders. But even though the theory contains some limitation, Keown et.al (2007) argues that there are still many individual investors and financial institutions who consider that dividends are important and it is, therefore, of 15 critical to comprise the theory even though it has some limitations. As we state in the section above, the bird in hand theory is the opposing view towards Modigliani and Miller’s dividend irrelevance theory and it says among other things that companies with greater earnings pay more dividends to its shareholders. As this view is the opposite view, compared to Modigliani and Miller’s, we think that it would be engaging to consider whether companies with a greater profit pay more dividends to its shareholders. 2.2.3. Signaling Theory The signaling theory of dividends has its origins in Lintner’s (1956) studies who reveals that the stock price of a company usually changes when the dividend payments changes. Even though Modigliani and Miller (1961) argues in favor of the dividend irrelevance they also states that in the real world disregarding the perfect capital markets, dividend supports an “information content” which may influence the market price of the stock. A number of researchers have thereafter been promoting the signaling theory and today it is seen as one of the most influential dividend theories. Bhattacharya (1979) introduces one of the most acknowledged studies regarding signaling theories which states that dividends may function as a sign of expected future cash flows. A raise in the dividends announces that the executives require higher cash flows in the future. The research is relied on the assumptions that outside investors have imperfect information respecting the corporate future cash flows and capital gains. Another important assumption is that dividends are taxed at a higher rate comparatively to capital gains. Bhattacharya (1979) discusses that under these circumstances, even though there is a tax disadvantage for dividends, firms would choose to pay dividends so as to send positive signals to shareholders and outside 16 investors. Baker (2009) affirms that a corporate origin of information like accounting data and future forecast reports are not totally trustworthy. These kinds of information do not thoroughly perform a company’s beneficial business occasions in the future. Provided that outside investors have imperfect information concerning the firm’s profit circumstances, the firm has to search other ways to persuade investors about future cash flows and earnings. Hence, favorable signals like raising dividends provide a favorable sign to investors. Although dividends have a higher tax rate comparatively to capital gains, outside investors are pleasing to pay at this rate for dividends in exchange for the effective signal dividends send as regards the value of the stocks. As a result of the signaling capability, dividends may be able to transform inefficient markets to perfect markets with full information efficiency. The signaling hypothesis is further developed by Miller and Rock (1985) who state that there is a high degree of information asymmetry between managers and outside investors. They further state that almost any company is able to pay small dividends to its shareholders regardless of whether its future outlooks are positive or negative. Consequently, if the dividend payments should be seen as a signal for profitable future business opportunities the dividend has to be enough large so that only companies with profitable future prospects can afford to pay it. Otherwise, companies with poor future prospects would just copy the signals and pay the same amount of dividends in order to send false signals to investors. Lots of researches have been handled to test if the signaling theory applies in the real world and there exist different opinions regarding the applicability of the signaling theory. Asquith and Mullins (1983) provide empirical evidence in favor of the signaling theory. They argue that an increase of dividend payments tends to increase the shareholders wealth. They also state that dividends contain information 17 which is not accessible in other sources of information for instance accounting data. But the signaling theory cannot be seen as applicable in all situations and a lot of researchers have found various drawbacks with the theory; for example, Pettit (1972) and Black (1976) state that the informational role of dividends are exaggerated and there exist less expensive way to signal the same information to shareholders. 2.2.4. Agency Theory The agency theory is one of the most respected dividend theories and it has been extensively debated among various scholars. One of the most influential studies regarding agency costs is presented by Jensen and Meckling (1976). The study provides a new view of the agency problem and most studies concerning agency costs use Jensen and Meckling’s research as a criterion. They describe the agency cost as a cost that arises between the principals (stockholders) and the agents (management), where the principals hire and delegate the agents with a power to maximize the wealth of the principals. They further state that only bonds and stocks can be applied as claims for the company. Hence, only shareholders and creditors can be seen as principals. Jensen and Meckling present a prominent research regarding agency costs and they provide a clear definition of what the agency cost is. But they do not provide a thorough corroboration regarding the effect of agency cost on dividend policies and many scholars have been trying to develop the theory. Easterbrook (1984) presents another study regarding agency costs and his result supports the findings made by Jensen (1976) and Rozeff (1982). Easterbrook organizes an investigation of whether dividend disbursement can be used to minimize the agency costs between investors and executives; he states that two causes influence the agency costs in a company, monitoring costs and the risk aversion preferences of 18 executives. The monitoring costs induce by the shareholders so as to manage the executives and avoid them from conforming their own personal agendas despite of maximizing the equity value of the shareholders. The second one of agency costs is the risk aversion preferences of managers. The issue arises because most of the shareholders have expanded portfolios and they are therefore only interested in systematic risk which can’t be removed through diversification. In difference to shareholders, executives usually have a large number of their personal wealth associated with the company. For that reason, if the company is loss or even goes bankrupt, the managers’ personal property turns into strongly influenced. The managers will be more risk averse in comparisons with the shareholders and they might reduce potential high value project because of their risk aversion preferences. According to Easterbrook (1984), these two causes of agency cost could be eliminated by paying dividends to shareholders. Nevertheless, Easterbrook further announces that dividends are worthless in themselves and companies should as a result only pay dividends so as to eliminate agency conflicts. Another theory that suggests the agency cost is the free cash flow theory (Jensen, 1986). The research explains that the agency costs increase while the free cash flow arises, because the shareholders have to boost the management to avoid the managers from engaging in enormous spending or losing investments, for example empire building. This can be made clearly by the positive interaction between the company size and the enumeration plan of management (Murphy, 1985). In order to prevent these kinds of conflicts between shareholders and managers, Jensen argues that companies should pay excessive free cash flow in form of dividends to shareholders. Otherwise the managers may follow their own personal interest instead of maximizing the wealth of the shareholders. 19 2.3. Determinants of dividend payout ratio and hypotheses 2.3.1. Financial leverage Financial leverage (sometimes referred to as gearing in the United Kingdom and Australia) is the application of borrowed money to boost production amount, and so sales and earnings. The larger amount of debt, the larger the financial leverage. An “unleveraged firm” uses only equity capital when in fact a “leveraged firm” uses a mixture of equity and various forms of debt. One of key motivation for this paper is the observation of Adedeji (1998) in UK that there is a positive reaction between dividend payout ratio and financial leverage over the period 1993 – 1996 in the United Kingdom. Moreover, Chang and Rhee (1990) also state a positive reaction because they explain that higher financial leverage exists in reaction to lowering shareholder tax rates, hence it makes the firm pleasing to pay higher dividends. On the opposite, point of view of Jensen (1992) is unfamiliar in that they believes making finance from equity is more pleasant to firms having high dividend ratios than from the debt, hence low scale of long-term debt over the book value of total assets frequently happen in these companies. Bebczuk (2005) and Kowalewski (2008) admit with the idea of Jensen (1992) in that they consider that firms with high leverage seem not want to compensate high dividends and take more loans with the intent of restricting default risk. Impartially, Al-Najjar and Hussainey (2009) announce there is no reaction between these two factors due to the statistically insignificant outcomes of hypothesis tests. Naceur’s (2006) results confirm that the total debt over the equity market value does not influence the dividend yield. 20 2.3.2. Investment Ardestani (2013) investigates the significance of investment opportunity set and corporate financing on dividend payout policy of Malaysian industrial products zone by selecting 62 listed companies on the main board of Bursa Malaysia. The result suggests that investment debt maturity and opportunity set are the components that significantly impact dividend payout policy of the sample firms. In most case, financial clinicians are arranged into three groups on the basis of their faiths on the influence of dividend policy on firm value. The first group considers that dividends contain information content that an increase in the dividend payout raises firm value (Lonie, 1996; McCluskey, 2006). The second is at the point of view that an increase in dividend has a tendency to diminish share price because (i) it suggests that firms have an insufficiency of positive NPV projects needing investment (Ghosh and Woolridge, 1988; Soter,1996) or (ii) it guides to higher taxation payments when the income tax is greater than that a capital gains (Lasfer, 1995; Bell and Jenkinson, 2002). The third group believes that dividend policy has no effect on firm value (Uddin, 2003; Kaleem and Salahuddin, 2006). 2.3.3. Risk Al-Najjar and Hussainey (2009) determine business risk as the chance of decrease in returns on investment due to exceptional circumstances. Under transaction cost theory, Rozeff (1982) offers that the transaction costs of external financing will be greater when the firm has greater operating and financial leverage, or more risks that can be calculated through the greater beta coefficient. Chang and Rhee (1990) propose the negative reaction that a firm with lower risk or more stability of earnings has more capability for remaining or paying out more dividends in the future. 21 Aivazain (2003), together with Li and Zhao (2008) recognize with the suggestion of Chang and Rhee’s (1990) in that they conclude that the lower the business risk the higher the dividends shell out. In addition, it has been argued that high-risk firms tend to have a higher volatility in their cash flows, than low-risk firms. Consequently, the external financing requirement of such firms will increase, driving them to reduce the dividend payout to avoid costly external financing (Higgins, 1972; McCabe, 1979; Rozeff, 1982). Jensen (1992) contends that greater systematic risk increases the uncertainty of the direct relationship between current and expected future profits. Hence, firms avoid commitment to pay large dividends, as the uncertainty about earnings increases. Mohd (1995) also reports an inverse relationship between the dividend ratio and intrinsic business risk, proxies by beta. He indicates that firms with unstable earnings pay lower dividends, in an attempt to keep the dividend payout stable and to avoid the high cost of external financing. In contrast, Mollah (2002) finds that firms listed on the Dhaka Stock Exchange pay a large dividend, even though the beta for their stock is high. 2.3.4. Corporate income tax Tax, in the real world, is a very important issue that needs to be concerned. Corporate income tax is expected to have a negative impact on dividend payout ratio. If the tax rate of a country is increased, ‘all things being equal’ there will be a reduction in the amount of distributable earnings left to be paid out as dividend hence the negative impact. Bhattacharya (1979) argues that under these circumstances (investors have imperfect information in relation to the company’s future cash flows and capital 22 gains; another important assumption is that dividends are at the higher tax rate compared to capital gains) even though there is a tax disadvantage for dividends, corporate would choose to pay dividends with the purpose to send positive signals to shareholders and outside investors. Tax-adjusted model (Miller, 1986) presumes that investors require and secure higher expected returns on shares of dividend-paying stocks. The consequence of taxadjusted theory is division of investors into dividend tax clientele. Modigliani (1961) argues that the clientele effect is responsible for the alterations in portfolio composition. Masulis and Trueman (1988)'s pattern forecast that investors with different tax liabilities will not be uniform in their ideal firm dividend policy. They conclude that as a tax liability increases (decreases), the preference for dividend payment also increases (decreases). Tax-adjusted model assumes that investors maximize after-tax income. As far back, Farrar and Selwyn (1967) conclude that in a partial equilibrium framework, individual investors choose the amount of personal and corporate leverage and also whether to receive corporate reimbursements as dividends or capital gain. 2.3.5. Liquidity The liquidity is the extent to which an asset or a security can be purchased or sold in the market without influencing the asset's price. Assets that can be easily sold or bought are known as liquid assets. This also is the ability to convert an asset into cash quickly also known as "marketability”. A weak liquidity position means less plentiful dividend owing to shortage of cash. Ho (2003) finds that more liquid firms in Japan have greater dividend payouts. Mehar (2002), notwithstanding, proposes there is an inverse reaction between liquidity position and dividend imbursements from his 23 research of companies on the Karachi Stock Exchange in Pakistan as companies with positive working capital will pay lower dividends. Myers and Bacon (2004) emphasizes that corporations are probably to lessen dividends to spread liquidity. Yet, several years later, Al-Najjar and Hussainey (2009) claim that paying lower or higher dividends does not base on a good or bad liquidity position. 2.3.6. Firm size Study by Lloyd, Jahera, and Page (1985), and Vogt (1994) suggest that firm size acts a role in expressing the dividend-payout ratio of firms. They discover that larger firms have a tendency to be more mature and hence get easier access to the capital markets, which eliminates their reliance on internally generated funding and allows for greater dividend-payout ratios. There is a fact that large firms will pay a significant amount of dividends to reduce agency costs (Ghosh and Woolridge, 1988; Eddy and Seifert, 1988). Eddy and Seifert (1988) also identify that large firms distribute a greater amount of their net profits as cash dividends than do small firms. Other studies also mention the positive association between dividends and firm size to transaction costs. For example, Holder (1998) affirms that larger firms find it easier to access to capital markets and find it easier to raise funds at lower costs, allowing them to pay higher dividends to shareholders. This demonstrates a positive association between dividend payouts and firm size. The positive relationship between dividend payout policy and firm size is also by an increasing number of other studies (Eddy and Seifert, 1988; Jensen et al., 1992) 2.3.7. Structure of asset Myers et al.(1984) presumes that companies possessing the most tangible assets tend to borrow more than those possessing the most intangible assets, thus they have 24 the capability to maintain less retained earnings and reimburse higher dividends following to pecking order theory. So, a positive connection between the tangible assets ratio and dividend payments is predicted. However, Aivazian’s (2003) outcomes points that inverse linking between tangibility of assets and dividend policy presents in listed companies in some emerging markets. The point of view of AlNajjar and Hussainey (2009) is indifferent in that they think that the much more tangibility the much more dependence on retained earnings due to lower financing from debts, then the lesser dividends are paid towards shareholders. 2.3.8. Sales Growth Sales growth may impact dividend payout ratios. Higher growth companies need greater requirement for external financing. Hence, to insure the access to external equity capital, the company may be motivated to set up a good reputation with stock holders through higher dividend payment (Laporta, R., F. Lopez-desilanes, A. Schiefer and R. Vishny, 1998). According to signaling theory, the high growth companies are smoother to pay dividends to their shareholders. Kania & Bacon (2005) research the affect of risk, liquidity, profitability, sales growth, and expansion on the dividend policy of a corporation by examining the financial data of over 10,000 publicly traded companies applying Ordinary Least Squares (OLS), the research makes a conclusion that the dividend payout ratio is significantly influenced by liquidity (current ratio), control (insider ownership), the profitability (return on equity), growth (sales growth), risk (beta), and expansion (growth in capital spending). Chang and Rhee (1990) also emphasize that companies having more opportunities for growth prefer retaining earnings to finance the development to 25 paying dividends, thus the more the growth opportunities, the lower the dividend reimburse. Growth rate is measured as the growth rate of sales (Rozeff, 1982; Mohd et al., 1995; Holder et al., 1998; Lloyd et al., 1985). Overall literature portrays a positive as well as a negative relationship between sales growth and dividend payout ratio. 2.3.9. Cash flow Cash flow is the movement of money into or out of a project, financial product, or business. It is usually calculated during a limited, specified period of time. The cash flow position of a firm acts an important role in determinant of dividend payouts. Alli (1993) argues that dividend payout ratio count more on cash flows, which indicates the corporate ability to carry out dividend payments, than on current profits, which are less strongly influenced by accounting practices. They claim that current profits do not really reflect the firm's ability to pay dividends. Lease (2000) indicates that companies with the higher cash flow are likely to have higher dividend payment. Amidu & Abor (2006) research outcomes present that, firms with high cash flow have the probability to pay high dividends to their shareholders. Nonetheless, Baker (2006) explains that most firms nowadays practice unmodified residual policy where the firms carefully manage their payout ratio and dividend stream after investment decision is made. While the firms may consistently have experienced low free cash flows, the dividend policy is not necessarily a corporate goal. 2.3.10. Market – to – book value 26 Market-to-book value indicates the market point of view of the value of equity in comparison with what shareholders have contributed to the firm since the day it was established. Omran and Pointon (2004) point that market-to-book value is a critical determinant that influence dividend payout ratio. Amidu & Abor (2006) research the determinants of dividend payout policy in Ghana, clinicians analyze the data in six years of listed companies on the stock exchange. Variables such as cash flow, corporate tax, profitability, market to book value, and sales growth are used. Results show positive reactions between dividend payout ratios and cash flow, profitability and tax while the relationship is adverse with the market to book value and growth. Gill, Biger, & Tibrewala (2010) indicate a positive relation between market to book value ratio and dividend payout ratio in the manufacturing sector but not in the service sector. 2.3.11. Ownership concentration Jensen and Meckling (1976) support for the viewpoint that large owner have a stronger incentive and more opportunities to monitor managers than small shareholders. Therefore, in case those large shareholders are not insiders, to set managers under their control and act on shareholders interest, they may force the firm to pay high dividend. Claessens and Djankov (1999) claim that the high concentration of owners reduce managers’ intentions to invest in low return project, hence more cash flows can be pay-out as dividends. Kouki and Guizani (2009) show that Tunisian firms having a high level of ownership concentration paying out more dividends. Ramli (2010) with the evidence of Malaysian listed companies finds a positive relationship between ownership concentration and dividend payment because controlling shareholders have greater influence over the dividend payout policy. 27 On the other hand, there are many studies which have been conducted to define this relationship and have the conclusion that the higher level of ownership concentration, the lower the dividend pay-out ratio. Based on agency cost theory, more concentration in ownership reduces agency conflicts. Hence, high dividend payment for the purpose of controlling managers’ action is not necessary. In fact, several studies suggest that firms with large concentrated ownership have low dividend pay-out ratio. According to Shleifer and Vishny (1986), concentrated ownership creates incentives for large shareholders to control the firm’s management with the purpose of protecting their investment. Ownership concentration controls the free-rider problem associated with ownership dispersion (dispersed ownership refers to the large number of small shareholders who own less than 5% of a firm’s stock). In addition, because large shareholders play an important role in the firm’s decision, managers will also give them preferences. Farinha (2003), Chen (2005) show that firms in US, UK and Hong Kong respectively, ownership concentration (especially concentrated insider ownership) have the negative effect on dividend policy. When interests of shareholders and managers are close, less dividend are paid out. The result of research by Maury and Pajuste (2002), Gugler and Yutoglu (2003) show that firms with the existence of high ownership concentration tend to pay lower dividends. Most recently, Harada and Nguyen in their studies about "Ownership concentration and dividend policy in Japan" in 2011 also show the negative effect of ownership concentration on the dividend payment. 2.3.12. Return on Equity The decision to make dividends payment starts with profits. Thus, it is logical to recognize profitability as a beginning factor, and the degree of profitability as one of 28 the most critical determinants that might affect firms’ dividend decisions. Profitability can be described as the ability of the firm to make a profit .The size of profit in a firm has been a factor of dividend policy representing for years. Directors more frequently do not suggest the payment for dividend when the firm has made remarkable profit require such payments. Dividend policy has been strongly and directly affected by profitability as among the major distinctions (Al-Kuwari, 2009). It is designed that money-making companies most tend to pay a dividend in comparisons with nonprofitable companies (Eriostis and Vasiliou, 2003; Ahmed and Javid, 2009). Accordingly, Pruitt and Gitman (1991) make to a close those current and previous years’ profits, the year-to-year and past years’ dividend are important determinants that impact dividend policy. The fraction dividend payout relies on the current earnings of the firm. They explain that, the higher the earnings, the more dividends will be made payment to the investors. They are speedy to compromise that profitability is supported by signaling theory as the company wants to push the standing of its performance. Nonetheless, Al- Kuwari (1998) remarks that executives arrange to pay a smaller amount of dividends now to handle with the changes in the future when a firm expects less cash flow in the future. In addition, Kowalewski (2007) notes that when a company has less investment opportunities it pays higher dividend from the earning profit. In his traditional research, Lintner (1956) points out that a company’s net earnings are the important factors of dividend changes. Accordingly, Al-Malkawi (2007) comes to the conclusion that profitability is a remarkable factor of the degree of dividends paid by firms. Reddy (2006) states that the dividends paid by companies are more profitable. Amidu and Abor (2006) support to the presence theories that decision for dividend payout of listed companies in Ghana is influenced by return on equity. They examine the relationship between the 29 number of company selected determinants and the dividend payout in Ghana. The sample includes of companies that have been listed on Ghana stock exchange during 1998-2003 and even though the sample just consists of 20 companies, they represent 76 percent of all listed firms in Ghana during the time period. And they give a positive reaction between the companies’ dividend payout ratios and profitability and cash flow. 2.4. Model for research According to these above previous researches we define dividend payout ratio as a function of financial leverage ,market-to-book value ratio, cash flow, liquidity, firm size, ownership concentration, sales growth, return on equity, corporate tax, risk, structure of asset, and investment. Based on the above definitions and researches, we obtained the following equations to analyze: div = β0 + β1finlev + β2invest + β3risk + β4tax + β5liq + β6lnsales + β7str + β8gro + β9cf + β10own + β11ROE + β12mtbv + ε Where, div, finlev, invest, risk, tax, liq, lnsales, str, gro, cf, own, ROE , and mtbv stand for dividend payout ratio, financial leverage, investment, business risk, corporate tax, liquidity, firm size, structure of ssset, sales growth, cash flow, ownership concentration, return on equity, and market-to-book value respectively. Name Abb Formula Dividend payout ratio DIV Dividend paid/Net income Financial leverage FINLEV Total liabilities/Total assets 30 (Current total asset – Previous total asset)/Previous total asset Standard deviation [Earnings before extraordinary items/ Total assets] Investment INVEST Risk RISK Corporate tax TAX Liquidity LIQ Firm size LNSALES Natural log of sales Structure of asset STR Fixed asset/Market value of equity Sales growth GRO Cash flow CF Market-to-book value MTBV Ownership concentration Return on equity OWN ROE Corporate income tax/EBT (Current asset – Current liability)/Current liability (Current sale – Previous sale)/Previous sale Cash flow form operation/ Total asset [(Shares outstanding*Share closing price)+Book value of total liabilities/Total book value of assets] Percentage of majority shareholders’ holdings of equity stock EBIT/ Market value of equity 2.5 Hypotheses Based on the theories and previous researches discussed in chapter two, the list below is the hypothesis for the research paper:  H1: There is a positive correlation between financial leverage factor and dividend payout ratio.  H2: There is a negative correlation between investment factor and dividend payout ratio.  H3: There is a negative correlation between business risk factor and dividend payout ratio.  H4: There is a negative correlation between corporate income tax factor and dividend payout ratio. 31  H5: There is a positive correlation between liquidity factor and dividend payout ratio.  H6: There is a positive correlation between structure of asset factor and dividend payout ratio.  H7: There is a positive correlation between sales growth factor and dividend payout ratio.  H8: There is a positive correlation between cash flow factor and dividend payout ratio.  H9: There is a negative correlation between market-to-book value factor and dividend payout ratio.  H10: There is a positive correlation between ownership concentration factor and dividend payout ratio.  H11: There is a negative correlation between firm size factor and dividend payout ratio.  H12: There is a positive correlation between return on equity factor and dividend payout ratio. 32 Chapter Three – Research Methodology This chapter concentrates on the details of the study undertaken to address the questions posed in chapter one. It defines the processes of building up the study conducted on the basis companies rating system, selecting the literature review, the data collection process including the tools and methodologies used in finally, data analysis and conclusion. 3.1. Problem statement Dividend policy has recently been seen as one of the major decisions of the company's financial management; which affects both the value of shares and the value of the business. In emerging market countries like Vietnam, due to asymmetric information, investors often rely on the dividend as a signal for the company's prospects for the future. Therefore, the dividend policy strategy is the important decision of the management. Established in July 2000, Vietnam stock market is typical emerging capital market that some characteristics can affect the dividend policy of the enterprise: asymmetric information, agency costs presence, emotional behavior of retail investors. In addition to emerging markets, the provision of the law does not complete due to the risk of policy changes, which may also affects dividend policy, such as changes in tax rules may change decisions payment of business. Listed companies in Vietnam stock market do not have much experience in managing finance status such as reinvestment or paying dividend, and investors have less acquirement in assessing the value of dividend. For instance, the company paying high dividends often means that the investors infer good prospects and vice versa. In 33 fact, a number of institutional investors favor companies with stable dividend policy to create a stable cash flow, and they are willing to pay a higher price for the shares it holds. This is what businesses need to be aware and should also aim to study the structure of shareholders to be able to establish a consistent dividend policy. However, the "process stability" can also lead to missed business opportunities good investment, which can increase the market price of the company in the future. In case if companies do not want to miss the opportunity, businesses will have to borrow, which means increasing financial risk, control issues change after a period of stable operation is not easy. Therefore, research needs to explore determinants affecting in dividend policy, particularly Vietnam market only. Although there are many researches on this topic, only just few of them can be applied for Vietnam due to different situations, for example, different cultures lead to “not-the-same” psychological behavior of investors which can be heavy-weighted components affecting dividend policy. The research will use quantitative test to come to the equation of determinants which are explored by qualitative method before. 3.2. Research objective The major target of this paper research is to reach the analysis and evaluation of the dividend payout ratio of listed companies on the HoChiMinh City stock market, which can be representative for Vietnam stock market. 3.3. Data Data for 5 years (2008-2012) was taken for 301 different companies listed in HOSE. The reason why I limit research only on the dividend policy of listed 34 companies on the HochiMinhCity stock market – HOSE - is that only these companies have transparent information of financial reports as well as dividend policy in asymmetric information market such as Vietnam today. Moreover, we can see that the listed companies are represented for the firms being aware of the importance and impact of dividend policy on the value of the business. The criteria for selection are the companies which match all for mutual five years for which data are obtained. The dividend paid each year constitutes cash dividends and capitalization in investors’ equity holdings. Dividends paid in cash and the dividends capitalized to shareholders equity are both taken as “dividends” as shareholders’ return is maximized with both ways. This is relied on the assumption that stakeholders can get the equivalent of the capitalized dividend in cash if they choose so. The weighted average number of shares outstanding during the year is taken to calculate Earnings Per Share (EPS) and dividends per share. It is important to take the average of the number of outstanding shares as this figure during a year varies day by day. The dividend per share, as well as Earnings per Share (EPS) figures are stated relative to par values of the respective companies for the reason the companies set varied par values for their stocks. The clearing share prices are obtained from the registry of buy/sell transactions of investors. Interviews are conducted to get the average clearing prices all over each year. For those stocks, the ownership transfer among investors is infrequent, the share premium (as set by management or the general meeting of shareholders) does addition to the par value to reach the potential clearing price (the price at which the shares were sold to new investors). 35 3.4. Research method Research design is the road map for researchers and it is a master plan specifying the approaches and procedures for collecting and evaluating information needed. (Zikmund, 1997, p.48). Based on observation, analysis and synthesis are keys. Based on the understanding of dividend policy theories and analysis, assessment of financial data and information on HoChiMinhCity stock market is built. The research paper is an experimental research. Experimental research methods are a class of research methods in which experimental observations or data are gathered so as to answer particular research questions. According to Singleton and Straits (1999), different research problems have different approaches. However, the two main approaches are qualitative approach and quantitative approach. Qualitative methods seem to be more applicable in the early stages of research and for theory building. Whereas, quantitative methods tend to be more appropriate when theory is well prepared and selected, for purpose of theory testing and analyzing. In reality, no research method is totally qualitative or quantitative (Yin, 1994). The research follows the quantitative approach. Data are not collected from directly respondent, instead published reports from HOSE are utilized. Data from year 2008 to 2012 are taken from this report. This follows the secondary data approach Secondary data: In addition to the methods and types discussed above, data collected in this study refer to secondary data. Data collected here is through financial reports. It cuts down the concern placed on respondents by contradicting the need to recruit further subjects and that it permits broad use of data from limited or unattainable respondents. 36 Furthermore, it has suggested that secondary analysis is more convenient approach for this research. 3.5. Data analysis procedure The financial reports are used to examine and compare variables of interest (dividend, financial leverage, and investment) among companies. The effect of the exogenous variable - market conditions is presumed to shake all studied companies in somewhat parallel manner. The confounding variable, which distorts the results if it had not been controlled, is the asymmetric information. 3.6. Data analysis types and method Simultaneity means that for instance in a natural structural macroeconomic model we clarify expenditure with income but income is also described by variables and not exogenous anymore. In this research, the method of multiple linear analyses is used to test the hypothesis. The following steps are an outline of the main elements of multiple regression analysis 1. Multiple regression analysis constructs on univariate and bivariate analysis, it is best to examine the data and the dissemination of the reaction and explanatory variables before attempting into multivariate analysis. This is then reflected by descriptive statistics and bivariate test in which the connections between the feedback and each particular explanatory variable are further analyzed. Once this information has been assembled, the relationship between some explanatory variables and the dependent variable may be further examined. Multiple regression will present the independent contribution of each explanatory variable to the forecast of the result while controlling for the domination influence of the other dependent variables. 37 2. Multiple linear regression draws a direct overview of the simple linear regression. The main dissimilarity in multiple regressions is the value of the dependent variable relies on some set independent variables instead of one. 3. In addition to, characterizing the relationship between the exploratory and dependent variables, multiple regression could be used to criticize the relative importance of different independent variables. This is done by measuring their relative t-ratios. 4. Categorical variables may be consisted in the regression equation by denoting them Yes/No values as 1/0. This is a strong point because one can look at attainable variations in the outcome in existence or absence of a categorical variable. If to be needed, a new categorical variable may be formed to serve as a numerical variable. For instance, a variable like age can be divided into two categories as young/old. Such variables are often mentioned to as dummy variables. The ability to merge continuous and dummy variables is one of the main benefits of multiple regression analysis in dealing with biomedical data. As with all statistical methods, multiple linear regression analysis bases on fundamental assumptions about the population from where the data have been acquired. The results of the analysis are only trustworthy, when these assumptions below are satisfied: a. The relationship between the independent variables and the dependent variable is linear. In another way, each increase by one unit in an independent variable is connected with a fixed increase in the dependent variable. 38 b. The regression equation explains the mean value of the outcome variable for a set of exploratory variables. c. The individual data points of Y (the outcome variable) for each of the independent variables are normally distributed about the line of regression. d. The variance points the regression line is alike for every explanatory variable. e. The independent variables are exploratory of each other, that is, knowing the value of one or more of the exploratory variables do not express anything about the others. This is often not the circumstance in real life because many variables are correlated. These correlations can lead to mistaken consequences. 3.7. Sample collection and data analysis In this chapter, the researcher outlines the research approach, including collection of data for quantitative analysis. This approach is used to generate new knowledge, new hypothesizes, or support for existing theories. The secondary analysis enables access to live data and the perspectives of specialists in the industries. 39 Chapter Four - Data Analysis In this section, the outcomes of the data analysis answering the research objective are presented. The purpose of this chapter is to present research results analyzing and checking the research framework in order to answer the main research objective “Based on the research model, examining the factors that determine the dividend payout ratio in HOSE’s market”. 4.1. Sample descriptive: As being planned in chapter three, the data to be collected are about from 301 listed companies on HOSE. However, there are some missing data because the companies’ listing points of time are different. For instance, the PTK code is listed in 2010 which makes us lack of data before listing point time. But our data are set under panel type; the problem of unbalanced data is acceptable. A summary of descriptive statistics is presented in table 4.1 Descriptive Statistics. The descriptive part in this research can’t draw an overall picture about the data because the data are calculated based on the real figures, in which variables don’t have a standard to have the comparisons. So, we couldn’t make the comparisons between variables just based on this descriptive table. This table shows the average indexes of variables computed from the 301 listed companies on HOSE. The average (mean) dividend payout ratio (calculated by dividend paid/net income) is .1048. This means, on the average, these 301 companies pay about 11 percent of their profit as dividend. 40 Table 4.1 Descriptive Statistic from 2008 to 2012 Variable Obs Mean Std. Dev Min Max div cf mtbv own finlev liq str gro invest tax ROE lnsales risk 1505 1505 1505 1505 1505 1504 1505 1500 1505 1505 1505 1505 1505 .1047748 .0407492 .8374722 44.61432 .4944507 1.566272 1.170243 .7923068 .7454818 .2434899 .1658753 13.08715 .0128186 .2989449 .1632981 .4393143 25.92744 .2252662 6.19228 1.746763 8.912381 13.18305 .1963395 .6580403 2.040132 .0776612 -.61659 -1.33499 0 0 0 -.88664 0 -.98349 -.6742211 -1.47536 -23.82661 0 -.7207 9.20525 1.18926 3.67161 100.01 .96436 157.9839 20.23692 228.7249 406.412 1.79302 3.43278 18.03944 .72214 Source: Table extracted from Stata software Receiving dividends from stock you own is great, but you want to make sure the company can afford to keep them coming. Companies pay dividends from their earnings, or profits. The dividend payout ratio measures the percentage of profits a company pays as dividends. A problem with the dividend payout ratio is that it might indicate that a company pays negative dividends since the denominator (net income) may be negative. When a company generates negative earnings, or a net loss, and still pays a dividend, it has a negative payout ratio. A negative payout ratio indicates that a company paid dividend even though it recorded a loss. A negative payout ratio of any size is typically a bad sign. It means the company had to use existing cash or raise additional money to pay the dividend. The value of this variable spreads from -0.6167 to 9.2053, the negative value denotes that even in a certain year the company gets lost from operation, it still pays dividend for investors. For instance, for a particular reason the company earns no profit in one year, it should make the investors feel at ease by paying dividend (even it’s 41 just a small amount). Al- Kuwari (2007) remarks that executives handle to pay smaller amount of dividends now to bear with the surprised changes in the future when a firm expects less cash flow at that time. Moreover, Kowalewski (2007) also marks that at whatever time a firm with less investment opportunities will pay greater dividend from the profit made. The company will keep amount of dividend in remarkable earning profit year to save for the bad performance year. The mean of cash flow variable is .0407 which tells that almost operating cash flow from companies comprise only 4% on total assets. This index is quite low, which shows that companies operate less effective in this period. With the market-to-book value of .8375, it shows the valuation of companies in comparison between market value and book value is not much different. The percentage of majority shareholders of listed company is 44.6%, these companies seem to have the focusing shareholdings. The financial leverage is to measure its ability to meet financial obligations; in this research, total liabilities take into account of nearly 50% of total assets. The liquidity ratio shows us the ability to convert an asset to cash quickly; in this case, this index of 1.566 is calculated based on current asset and current liability. Structure of asset 1.17 which means the assets are invested 1.17times more than company’s equity. The speed of sale growth is 79% on average, this index is quite good in the period of crisis economy from 2008 to 2012. Together with the sale growth, the investment on asset increases 74.5%; the minimum of this index is negative is because some companies reduce the value of asset comparatively with previous year. The tax rate is 24.34% which is the same with corporate income tax rate of Vietnam in this period. We can also find the negative tax index, it is because the company has to pay tax although it is loss. It is normal, for 42 instance, the total operating profit in a certain year of company is negative, but in this year, the company carries out the transfer land use right, the company still has to pay income tax from the profit for this transaction. The amount of net income returned as percentage of shareholders equity is 16.59%. The table shows the average firm has a size of 13.09 as measured by logarithm of its sales. 4.2. Panel data models 4.2.1. Fixed effect – Hausman test: Panel data may be group effects, time effects, or both. These effects are either fixed effect or random effect. A fixed effect model assumes differences in intercepts across groups or time periods, whereas a random effect model explores differences in error variances. (UITS Research Technology). In this research, we will analyze the impacts of independent variables on the dividend payout ratio that varies over time. Therefore, fixed effect is an appropriate choice. However, Hausman test will be used to test for sure. Fixed effect explores the relationship between predictor variables of year and outcome variables within an entity (company). Each entity has its own individual distinction that may or may not affect the predictor variables. The “time” could influence the dividend payout ratio of each company, for instance, due to the crisis economy in 2008, companies tend to carefully finance the operations which lead to the amount of paying dividend. 43 To decide between a random effects and fixed effects model, researchers often rely on the Hausman (1978) specification test (e.g., Greene 2008, 208-209). The Hausman test is designed to detect violation of the random effects modeling assumption that the explanatory variables are orthogonal to the unit effects. Under the null hypothesis of orthogonality, H is distributed chi-square with degrees of freedom equal to the number of regressors in the model. A finding that p < 0.05 is taken as evidence that, at conventional levels of significance, the two models are different enough to reject the null hypothesis, and hence to reject the random effects model in favor of the fixed effects model (Clark & Linzer, 2012) To serve Hausman test purpose, we need to analyze the collected data by fixed effect method and random effect method: Table 4.2 Hausman test result (B) (b – B) Sqrt (diag (V_b-V_B)) random Difference S.E. risk -.0914571 .2560103 .1588243 cf .0460307 .0930144 .067468 mtbv .055034 -.0047701 .0258131 own -.0001974 .0001188 .000675 finlev .1207972 .3124443 .1949434 liq .0037191 .0087102 .0078023 str -.0147471 .029232 .0285367 gro -.0006547 .0010724 .0019434 invest -.0209544 -.0055156 .0232545 tax -.319997 -.0709859 .1394903 ROE 1.013069 .5432001 .105988 lnsales .001446 -.0271077 .0129637 b = consistent under Ho and Ha; obtained from xtreg B = inconsistent under Ha, efficient under Ho; obtained from xtreg Test: Ho: difference in coefficients not systematic chi2(12) = (b-B)'[(V_b-V_B)^(-1)](b-B) (b) Fixed .1645533 .1390451 .0502639 -.0000786 .4332415 .0124293 .0144849 .0004177 -.02647 -.3909829 1.556269 -.0256616 44 = Prob>chi2 = 35.15 0.0004 Source: Table extracted from Stata software Based on the Hausman test, we come to the conclusion that fixed effect medoling will used in this study. 4.2.2. Checking correlation Correlation is a statistical method that can point out whether and how closely pairs of variables are related. Even though this correlation is fairly clear, the data may consist of undoubtful correlations, you may also distrust there are correlations, but don't know which is the most closest. A knowledgeable correlation analysis can lead to a better understanding of the data. The primary outcome of a correlation is called the correlation coefficient (or "r"). It covers from -1.0 to +1.0. If r is near to 0, it means it exists no relationship between the variables. The closer r is to +1 or -1, the more strongly the two variables are related. If r is negative it means that as one becomes larger, the other reaches smaller (often called an "inverse" correlation). If r is positive, it shows that as one variable becomes larger the other reaches larger. According to Pearson correlation, if the value of r is higher than +.7 or lower than .7, it shows a very strong relationship. If the correlations between independent variables are too high which means the correlation “r” is higher than +.7 or lower than -.7, it could cause multi-collinearity problem which needs to be excluded in the multiple linear regression. This can produce problem in interpreting the coefficients of variables as several are providing duplicate information. 45 Table 4.3 shows the correlation amongst predictors. No correlation has absolute value larger than 0.7 suggesting that multi-collinearity is not a major problem, no variable is omitted due to suspected of multi-collinearity. However, since the data is analyzed from the actual financial statements, it can’t avoid the problem of imperfection. Hence, all of variables are qualified for analyzing in the next step. Although some index is quite high which nearly reaches the milestone of 0.7, these high one are almost unreliable (not denoted by star). And the data in this research is calculated based on real numbers, not like the standard number in questionnaire research type, results in table 4.3 is acceptable. Furthermore, to make sure for the correlation between independent variables, we process the test of VIF later. 46 Table 4.3 Correlation matrix div risk cf mtbv own finlev liq str gro invest tax ROE lnsales div**** 1.0000* -0.0485* -0.0148* 0.0438* -0.0142* 0.0066* -0.0148* -0.0281* -0.0124* -0.0072* -0.0359* 0.0475* 0.0568* risk**** 1.0000* 0.0693* 0.1592* 0.0313* -0.0671* 0.0096* -0.0436* 0.0539* 0.0233* 0.5100* 0.0309* 0.0535* cf**** 1.0000* 0.1994* 0.0430* -0.1718* -0.1309* 0.0145* -0.0266* -0.0760* 0.0993* -0.0074* 0.1219* mtbv*** 1.0000* 0.1935* -0.0187* 0.0292* 0.0335* -0.0226* -0.0115* 0.2087* 0.0252* 0.3002* own*** 1.0000* 0.0996* -0.0034* 0.1188* -0.0275* -0.0320* 0.1033* -0.0093* 0.2155* finlev*** 1.0000* -0.2905* 0.4259* -0.0449* -0.0123* 0.2649* 0.0434* 0.4006* liq***** 1.0000* -0.0917* 0.0172* 0.0083* -0.1103* 0.0037* -0.2382* str**** 1.0000 -0.0139 0.0068 0.0266 0.0321 0.2022 Source: Table extracted from Stata software gro*** 1.0000* 0.1164* -0.0245* 0.0089* -0.0337* invest* 1.0000* -0.0301* -0.0009* -0.0745* tax**** 1.0000* 0.0205* 0.3108* ROE*** 1.0000 0.0276 lnsales** 1.0000 46 4.2.3. Pooled Ordinary Least Square analysis The Ordinary least square (OLS) is applied to draw a general overview for the research model. This is a pooled regression model as we pool all the observations in OLS regression. The model is implicitly presuming that the coefficients (including the intercepts) are alike for all of the individuals. In other words, this step is analyzed according to cross-sectional type data. Since there is no variable rejected in the correlation testing step, all of factors are keeping for OLS. Table 4.4 presents the results of the OLS regression analysis being conducted in this study. The model regresses the dividend payout ratio (the dependent variable) on the business risk (risk), cash flow (cf), market-to-book-value (mtbv), ownership concentration (own), financial leverage (finlev), liquidity (liq), structure of assets (str), sales growth (gro), investment (invest), corporate income tax (tax), return on equity (ROE) and firm size (lnsales) , which are independent variables. The pooled model qualified the data well at the significance level (F = 10.79, p < 0.00). The value for R-squared in table 4.4 shows that 12.27 percent of the total systematic variations in dividend payout ratio can be explained by the variations in dependent variables. This points out that about 87.73 percent of the systematic variations in dividend payout can be accounted for by others factors outside the model. According to the result from the table below, five independent variables which have the significant correlation with the depended variable ; they are : market-to-book value, financial leverage, structure of asset, corporate income tax, and return on equity. Amongst significant relationship variables, return on equity has the most impact on the 47 dividend payout ratio with the coefficient index is around 1.Structure of asset and corporate income have the reverse influence on dividend, which means the increasing in the amount of two variables makes the amount in dividend decrease, and vice versa. In contrast with that, market-to-book value, financial leverage, and return on equity impact similarly on dividend payout ratio. Table 4.4 OLS regression result Source SS Df MS Model 14.2207049 12 1.18505874 Residual 101.677929 926 .109803379 Total 115.898634 div Coef. risk -.0914571 cf .0460307 mtbv .055034 own -.0001974 finlev .1207972 liq .0037191 str -.0147471 gro -.0006547 invest -.0209544 tax -.319997 ROE 1.013069 lnsales .001446 _cons -.0567861 938 .123559311 Std. Err. t .1913705 -0.48 .077334 0.59 .0270115 2.04 .0004649 -0.42 .0766421 1.58 .0063355 0.59 .006579 -2.24 .0010949 -0.60 .0238922 -0.88 .0870585 -3.68 .1038904 9.75 .0064282 0.22 .0740258 -0.77 Number of obs = F(12,926) = Prob > F = R-squared = Adj R-squared = Root MSE = P>|t| [95% Conf. 0.633 -.4670273 0.554 -.1065233 0.042 .0020232 0.671 -.0011099 0.015 -.0296151 0.557 -.0087146 0.025 -.0276586 0.550 -.0028035 0.381 -.0678436 0.000 -.4908518 0.000 .8091809 0.822 -.0111694 0.443 -.202064 939 10.79 0.0000 0.1227 0.1113 .33137 Interval] .2841131 .1985847 .1080448 .000715 .2712095 .0161527 -.0018356 .0014941 .0259347 -.1491422 1.216957 .0140615 .0884917 Source: Table extracted from Stata software 4.2.4. Testing multi-collinearity Multi-collinearity is a widespread obstacle in regression analysis. It is common that we need to routinely examine in each model we analyze, as its existence will generate results that guide us to make erroneous conclusions with our hypothesis tests. 48 Multi-collinearity is a concern for multiple regressions, not for its existence, but for its degree. For severe degree of multi-collinearity, the regression model measures of the coefficients become changeable and the standard errors for the coefficients can get uncontrollably inflated. Formally, variance inflation factors (VIF) measures how much the variance of the estimated coefficients is raised up over the case of no correlation among the X variables. If there is no-correlation between two X variables, then all the VIFs will be 1. Variance inflation factor (VIF) is just the complementary of a tolerance value; hence low tolerances correspond to high VIF. Freund and Littell (2000) states that VIF show how multi-collinearity has boosted the instability of the coefficient estimates. In another way, VIF lets us know how "inflated" the variance of the coefficient is, in contrast with what it could be unless the variable is correlated with any other variable in the pattern (Allison, 1999). Table 4.5 VIF result Variable VIF 1/VIF finlev 2.50 0.399671 tax 1.75 0.571894 lnsales 1.47 0.681422 risk 1.46 0.683964 liq 1.45 0.689318 invest 1.33 0.752342 str 1.33 0.753203 cf 1.26 0.794458 gro 1.23 0.812044 mtbv 1.20 0.831315 own 1.10 0.909068 ROE 1.07 0.938347 Mean VIF 1.43 Source: Table extracted from Stata software 49 The VIF test tells us more about the multi-collinearity problem. If any variable exists the VIF figure of over 10, it proves there is problem of multi-collinearity in the research. In this study, the VIFs for all factors are qualified.. In table 4.4, none of predictor is suspected of causing multi-collinearity since their tolerance values are smaller than 1 or VIFs are larger than 10. Note that the 1/VIF is nothing other than the tolerance value. 4.2.5. Linear regression with large dummy variables set In the previous part, pooled OLS measure is still densely biased because of inaccessible heterogeneity. This is because of the fact that pooled OLS also bases on a between comparison. In comparative with the cross-sectional, OLS the bias is lower, nonetheless, because pooled OLS also takes concern of the within variation. Hence, panel data alone does not solve the problem of inaccessible heterogeneity. The collected data are set under panel type which contains observations of multiple phenomena acquired over multiple time periods for the same units or individuals. To promote the strength of panel data, we’ll continue analyzing linear regression with large dummy variables set. Multiple linear regression analysis allows the researcher to shape a data set for forecasting an outcome variable from one or more predictor variables. The model is symbolized by the regression equation, which is rigorously valid only within a range of data acquired. Trying to predict the result outside the range of the data can be intensely misleading, and is not well-considered. 50 When several independent variables are accountable for a dependent, each subject contributes a unique combination of these variables. A new subject on which an investigators wishes to predictably draw the outcome has also his own unique mixture of variables. The calculations may or may not fall within the range of the authentic study. It is as a result that prediction on the explanation of a given model can only be a guide and is not infallible. OLS is the commonly method for analyzing in economic research, however, because there are numbers of firms who don’t pay dividend (dividend payout ratio equals 0), OLS may be biased. Table 4.6 Linear regression with large dummy variables set result Linear regression, absorbing indicators div risk cf mtbv own finlev liq str gro invest tax ROE lnsales _cons company Coef. Std. Err. .1645533 .2459106 .1390451 .1017779 .0502639 .0369443 -.0000786 .0008105 .4332415 .2071253 .0124293 .0099382 .0144849 .0289577 .0004177 .0022057 -.02647 .0329679 -.3909829 .1625894 1.556269 .1467538 -.0256616 .0143081 .023404 .1606279 F (275 , 651) = T 0.67 1.37 1.36 -.010 2.09 1.25 0.50 0.19 -0.80 -2.40 10.60 -1.79 0.15 1.077 Number of obs = 939 F(12,651) = 10.88 Prob > F = 0.0000 R-squared = 0.3970 Adj R-squared = 0.1311 Root MSE = . 32766 P>|t| [95% Conf. Interval] 0.504 -.3183205 .647427 0.172 -.0608074 .3388977 0.174 -.0222805 .1228084 0.923 -.0016701 .0015128 0.037 .0265272 .8399058 0.212 -.0070856 .0319441 0.617 -.0423769 .0713466 0.850 -.0039134 .0047488 0.422 -.0912062 .0382662 0.016 -.7102458 -.0717201 0.000 1.268101 1.844437 0.073 -.0537572 .0024339 0.884 -.2920074 .3388153 0.229 (276 categories) Source: Table extracted from Stata software 51 As discussion above, fixed effect is used to analyze linear regression in this paper. The result of this step is showed in Table 4.6 above. The dummy variables in this case are the period time 2008, 2009, 2010, 2011, 2012. The significant level shows the second model fits the data better (F = 10.88, p < 0.00). Going by the value of R-squared of 0.3979, it shows that 39.8 percent of the total systematic variations in dividend payout ratio can be explained by the variations in dependent variables. Simultaneously, it means that about 60.2 percent of the systematic variations in dividend payout can be accounted for by others factors outside the model. This test comes to the conclusion that there are only three exploratory variables which have the significant influence on our dependent variable. The result in this model is a little bit different from the previous model. Five factors are proven in OLS test to have significant relationship with dividend payout ratio. Meanwhile this step continues deleting two more variables. This analysis tells us that it can explain the characteristic of dependent variable based on four significant variables up to 39.8 percent, which is higher than OLS analysis a lot. Therefore, we can conclude that with the collected data as type of panel data, the result pointed out from linear regression with large dummy variables set is more significant than the simple cross-sectional regression. 4.2.6. Linear regression with lagged dependent variable If it exists autocorrelation in the model, it is necessary to solve it. There might be panel specific autocorrelation or there might be common autocorrelation across all panels. There are supplies for identifying the type of autocorrelation. An autoregression 52 on lags of the residuals, alternatively, may indicate the existence or omission of autocorrelation and the need for dynamic panel analysis. If there is autocorrelation from one momentary period to another, it is possible to examine the “differences in differences” of these observations, applying the first or last as a foundation (Wooldridge, 2002). Blundell, Bond and Windmeijer (2000) recommend lagged dependent variables into their model to make narrative for dynamic effects. The lagged dependent variables can be announced to either fixed or random effects models. Their admittance assumes that the number of material consideration is greater than the number of repressors in the model. If autocorrelation consists across these observations, the model might be first partial differenced to manage for the autocorrelation effects on the residuals (Greene, 2002). Even if one presumes no autocorrelation, obstacles from the correlation of the lagged endogenous and the disruption term may bother the analysis. Especially, bias can result when the sample is finite or small. If one consumes general methods of moments, with instrumental variables, the usage of the proxy variables or instruments may prevent problems with correlations of errors. Besides, there are a large number of instruments implemented by lagged variables. Assume your dependent variable is expenditure. If expenditure today has an effect on the expenditure in future time points, then there will be correlation in the discovered values of expenditure consumption (called autocorrelation). To diminish this auto correlation, the lagged values can be fitted to the model. 53 For dynamic panels, with lagged dependent variables, with greater numbers of occasional conditions, they are able to conduct some missing data and they can accomplish gains in effectiveness as long as there are three or four periods of data (Greene, 2002). Let imagine that if you are an investor, each year your receive, let say, 20% dividend payment, in one year you only receive 5% dividend payout only, you would make a question and feel anxious about the performance of the company, even though it was due to crisis economic. Or in certain year, you receive a remarkable big amount of dividend than previous year, and received a small amount the next year, it makes you feel something wrong about the fluctuation about the company’s performance. Therefore, linear regression with lagged dependent variable is necessary to analyze to find out more, since the dividend payment in the past can influence on the current dividend payment. And the analysis with additional variable of lagged dividend payout ratio will let us a deep view on the picture of dividend payout ratio of companies in emerging market like Vietnam. Table 4.7 Linear regression with lagged dependent variable Linear regression, absorbing indicators div div L1. risk cf mtbv Coef. Std. Err. t -.0363538 .4284491 .2010712 .0456325 .1044983 .3310115 .1400582 .0525864 -0.35 1.29 1.44 0.87 Number of obs = F(13,479) = Prob > F = R-squared = Adj R-squared = Root MSE = P>|t| [95% Conf. 0.728 0.196 0.152 0.386 -.2416855 -.221965 -.0741333 -.057696 761 9.86 0.0000 0.4320 0.0988 . 36271 Interval] .1689779 1.078863 .4762756 .1489611 54 own finlev liq str gro invest tax ROE lnsales _cons company -.0002736 .0013776 .8454406 .2899535 .0160338 .0130276 .0191707 .0357782 .000991 .0025772 -.0199208 .0417216 -.6401966 .2567477 1.972619 .1936898 -.0139583 .0545511 -.3489179 .7321752 F (268 , 479) = -0.20 2.92 1.23 0.54 0.38 -0.48 -2.49 10.18 -0.26 -0.48 0.893 0.843 0.004 0.219 0.592 0.701 0.633 0.013 0.000 0.798 0.634 0.849 -.0029805 .0024333 .2757026 1.415179 -.0095645 .041632 -.0511309 .0894724 -.004073 .006055 -.1019008 .0620591 -1.144688 -.1357056 1.592033 2.353206 -.1211474 .0932308 -1.78759 1.089754 (269 categories) Source: Table extracted from Stata software Table 4.7 shows that F-statistic is significant ( F = 9.86, p< 0.05) at 95 percent confidence interval suggesting the model is useful to determine the variation in the criterion dividend payout. With R-squared index of 0.4320, this analysis shows that 43.2 percent of the total systematic variations in dividend payout ratio can be explained by the variations in dependent variables. Simultaneously, it means that about 56.8 percent of the systematic variations in dividend payout can be accounted for by others factors outside the model. 4.3. Discussion: The three analysis steps above fit the data well with the significant level below .05. The OLS method ignores time difference so it reports misleading the intercept. Linear regression with large dummy variable set considering the fixed group effect seems to be better than the simple OLS. The last step seems explain the almost characteristic for the collected data. 55 To sum up, among twelve tested variables, there are three variables which significant affects dividend policy of firms listed on the HOSE in the sample. They are financial leverage, return on equity, and corporate income tax; and based on the result above, return on equity is suggested to be the most important predictors to anticipate dividend payout pattern. Table 4.8 Result Comparison Items OLS L/R w. large L/R w. lagged dummy variable set dividend Intercept -.0568 .0234 -.3489 p-value of F-test .0000 .0000 .0000 R2 .1227 .3970 .4320 SEE (square root of MSE) .3314 .3277 .3627 Source: Author’s regression result comparison All of three models in this research point the significant reliable when all of index for model is in the threshold. The OLS step only allows us to keep five predictor variables for the study. However, the R-squared index is quite low with only 12.85 percent, and as explained in previous part, the result of OLS is biased and inconsistent. For this reason, the outcome in this step can only be seen as a reference for an overview about the data purpose. The rest two analysis steps are proved by other researchers to be enable to exploit all aspects of panel data type; and the R-squared indexes help us better 56 understanding about the main objective from this research. The comparison between empirical evidence and results concerning the determinants of dividend policy in Vietnam in this study is summarized in table 4.9: Table 4.9 Comparison between empirical evidence and results in the study Variables Proposition Financial Leverage Hypotheses Results Sig. Hypotheses Positive/ Negative Positive Positive level Yes status Accept Investment Positive/Negative Negative Negative No Reject Risk Negative Negative Positive No Reject Corporate Tax Negative Negative Negative Yes Accept Liquidity Positive Positive Positive No Reject Structure of Asset Positive Positive Positive No Reject Sales Growth Positive/Negative Positive Positive No Reject Cash Flow Positive/Negative Positive Positive No Reject Market-to-book Negative Negative Positive No Reject value Ownership Positive/Negative Positive Negative No Reject concentration Firm size Positive/Negative Negative Negative No Reject Return on equity Positive Positive Positive Yes Accept Source: Author’s comparison on empirical evidence and analyzed result 57 Amongst three of final significant influence on dividend payout ratio, the coefficient of return on equity is 1.97 which shows that this variable has the most positive impact on dependent variable. . The slope coefficient of this variable is 1.97, suggesting that a 1 unit increase in corporate income tax would increase 1.97 units in dividend payout ratio (ceteris paribus). In adverse, the slope coefficient of corporate income tax is -0.64, it means a 1 unit increase in corporate income tax would decrease 0.64 units in dependent variable. Market-to-book value is used as proxy for the firm’s future prospects and investment opportunities is found to have insignificant with dividend payout ratio. This measure is consistent with the findings of studies of Collins (1999) and Zeng (2003) implies that emerging market has different distinctions that lead concentration. If we consider the percentage of majority shareholders, tax-based theory has better discussion to explain the negative sign between dividend payout and ownership concentration because of dividend preference (Short, 2002). However, it is then remarkable that ownership concentration, which Jensen et al. (1992) regard as an another control tool, has a meaningless affect on dividend policy, as described by Renneboog and Trojanowski (2006) for the UK and Fenn and Liang (2001) for the US. Fenn and Liang (2001) for the US and Renneboog and Trojanowski (2006) for the UK. relied on the insignificant regression results, this factor is not the main concern of the firms to pay dividends. Alli (1993) also reveals that dividend reimbursement depends on cash flows, which demonstrate the company’s ability to cover dividend payments, than on current earnings, which are less strongly affected by accounting practices. Green (1993) questions the irrelevance problem and examined the relationship between the dividends and 58 investment. Their study shows that dividend decision is taken along with investment and financing decisions. The results however do not support the views of Miller and Modigliani (1961). Dhrymes and Kurz (1967) and McCabe (1979) find that the firm’s investment decision has the connection with its financing decision. Higgins (1972), and Smirlock and Marshall (1983) document no interdependence between investments and dividends. Mollah (2002) finds that firms listed on the Dhaka Stock Exchange paid a large dividend, even though the beta for their stock was high. He then argued that in an emerging stock exchange, the dividend might not be the most appropriate tool to convey correct information about transaction costs to the market. This study uses beta as a common proxy for firm business risk, which represents a firm’s operating and financial risk (Rozeff, 1982; Saxsena, 1999; Manos, 2002). The study of Abdul and Haruto (2012) examined the determinants of dividend payout ratio of the companies of Karachi Stock exchange 100 index for the year 2009, author also find that liquidity is found to be insignificant in the study. Based on these above results, we can conclude that our Hypotheses H2, H3, H5, H6, H7, H8, H9, H10, and H11 are not supported due to insignificant relationship to dividend payout ratio with 95 percent confidence interval. Hypothesis H1, H4 and H12 are partially supported by which financial leverage and corporate income tax are positively related to dividend payout with 95 percent confidence interval. 4.4. Answer research questions 4.4.1 Question 1 59 There is no doubt that several studies have been conducted (and is still on-going) on the examination of determinants influencing on dividend payout ratio, but outcome of these studies are mixed. This research paper is an exertion to expose the insight dynamics for factors of dividend payout with reference to listed companies on HOSE. In light of earlier literature, key independent variables are recognized to attach their associate and effect on determination of dividend payment. This research study investigates the connection between dividend payout ratio and other predictors calculated based on financial statements of listed companies. As being predicted, the result of this study paper shows that financial leverage and return on equity are statistically significant positive relationship with dividend payout ratio, suggesting that highly financial leveraged and highly return on equity firms pay higher dividend to their shareholders. Meanwhile, corporate income tax has a significant negative impact on dividend payout ratio. This result is consistent with some of previous researches. 4.4.1.1. Financial leverage Belden et al. (2005) finds out, when he checks 524 large American firms in the list of Forbes 500 from the period of 1998 to 2000, that debts are still existed in companies implementing dividend policy to manage agency problems. Higher financial leverage causes lower shareholder tax rate, hence it makes the firm pleasing to pay higher dividend. Moreover, the stock market in Vietnam is quite young, investors tend to be influenced by “crowded psychology”, and they become more sensitive with the debt 60 status of companies, so the debt level increase should come together with the increase in dividend payout for maintaining the share price as well as the funding from investors. In the crisis period started from 2008, the debt interest in Vietnamese bank could reach up to 20%, together with the governmental fiscal tightening package which make companies meet obstacles in accessing to capital from bank or government, hence the firms tend to search capital source from the stock market. The increase in amount of dividend payout would attract more investors, a retain existing investors for financing purpose. 4.4.1.2. Return on equity A recent research study of Denis and Osobov (2008) records the ratio of retained earnings to equity as a significant determination influencing dividend policies in the most six developed financial markets (the US, Germany, Canada, Japan, the UK, and France,). The return on equity tells common shareholders how effectively their money is being employed. Firms that have the experience of earning volatility find it difficult to make dividend payment, thus such companies will have high probability to pay less or no dividend. These results get the same point of view with previous empirical studies. In a research that studies whether dividend payout policy effects firm performance in the Ghana Stock Exchange, Amidu (2007) reveals that dividend payout ratio influences firm performance especially the profitability. The results come to the conclusion there is a positive and significant relationship between return on assets, return on equity, growth in sales and dividend policy. This proves that when a firm has a certain policy to pay dividends, its profitability is considered. The results also state a statistically significant relationship between profitability and dividend policy. Mizuno (2007) gets along to the 61 fact that firms should pay dividends to shareholders if they cannot identify appropriate investments which would bring higher returns than those expected by the shareholders. 4.4.1.3. Corporate income tax Corporate income tax is expected to have a negative impact on dividend payout ratio. If the tax rate of a country is increased, ‘all things being equal’ there will be a reduction in the amount of distributable earnings left to be paid out as dividend hence the negative impact. Dividends are currently paid out of after-tax amount, and interest charges from before-tax amount. Permission for firms to deduct dividends as they do interest charges would make dividends less costly to pay than before and would thus tend to increase the payout ratio. The finding in this result is consistent with existing literature. Tax, in the real world, is a very important issue that needs to be concerned. Bhattacharya (1979) argues that under certain circumstances (outside investors have imperfect information regarding the company’s future cash flows and capital gains, another important assumption is that dividend amounts are at a higher tax rate compared to capital gains) even though there is a tax disadvantage for dividends, companies would choose to pay dividends in order to send positive signals to shareholders and outside investors. 4.4.2. Question 2 In general, the value R-squared of final model in this research is average which endorses that more than 43% of the variation in the outcome variable is made clear by the exploratory variables of the model. Less than 57% variation in the dependent variable remains unexplained by the independent variables of the study. The result from analysis 62 step 2 is consistent with the result from step 3 which shows there are three factors being determinants of dividend payout ratio. The most critical point is that among examined factors to the sample, return on equity is the major determinant of dividend payout ratio in Vietnam. The variable of financial leverage has the second positive impact on dividend payment, and the last one corporate income tax makes the least negative influence on dividend ratio. For the reason, investors who take care of business environment in Vietnam can rely their expectations on the return on equity of a firm in the future to examine whether they should buy, hold or sell their shares. The more return on equity a firm can achieve, the more dividends securities holders are able to receive. The level of debt in one company increase which means the risk increase to, normally raise in risk will lead to rise in money (or amount of dividend). As explained above, in the current status of lacking of debt source, companies try to cover the dividend to attract new investors as a solution for capital financing. Therefore, with the current status, investors can base on the level of debt as a second source of choosing investment chance, but the shareholders should be careful when relying on this factor, because ability to pay debt is an important factor too. In conclusion, with proposed 12 hypotheses in chapter three, nine of them are rejected for insignificance in the analyzing progress. These results will be discussed further in next chapter. We could find out the theories as well as current event or status to explain clearly for the result testing in this chapter. For summary, this chapter shows that progress of analyzing data collected as well as explanation for phenomena happened in the progress of analyzing. According to the 63 result, the suggested model in chapter is explored by the listed companies. The results shows that in suggested 12 hypotheses, only 3 of them are accepted based on the significant level. 64 Chapter Five – Conclusion In this chapter, accomplishments and limitations to gain objectives and contents stated in chapter one will be evaluated. Moreover, in chapter four, data collected are analyzed and explained for what mentioned in chapter one. The result shows the data has the meaning and reasonably validity. In this chapter, the summary of what has been done will be summarized and from that, some recommendation will be suggested. 5.1. Conclusion This research uses multiple linear regressions in Stata software to analyze the determinants of dividend payout ratio in Vietnam based on a sample of 301 companies listed on the HOSE within the period of 2008 to 2012. It is the contribution to current Vietnamese literature by checking factors belonging to firms’ characteristics (profitability, firm size, financial leverage, liquidity, asset structure, business risk and growth opportunities) but also related to corporate governance that involves ownership concentration. Multivariate Regression Analysis is described as the most suitable instrument for econometric analysis of the data. The descriptive statistics does not make sense in this study because the independent variables are not standardized, variables can’t be compared with each other. The Hausman test confirms the fixed effects characteristics of collected and analyzed data. Meanwhile when the premises required to be fulfilled for OLS are tested, the data are pointed to be free of multi-collinearity and free of autocorrelation. After using three regressions method for analyzing, the results reveal that 65 financial leverage, corporate tax and return on equity have significant relationships with dividend payout. The rest of the nine explanatory variables are found to be insignificant reaction with dividend payout ratio in context of Vietnamese markets, especially with the chosen proxy of variables in this study. Actually, after analyzing the first step OLS, there are five exploratory variables to be proved to have the influence on the chosen outcome variable. However, the result from OLS is not reliable in the case of panel data like in this research. The second step of linear regression with large dummy variables set is used to analyze with the purpose of exploiting the characteristic of panel data. This step is not only cross-sectional analyze as OLS, it includes the time-series factor for more reliable result. This step reduces one more variable of structure of asset, and the index R-squared is increased remarkably. The last analyze method, linear regression with lagged dependent variable is used to test deeply whether the past record of dependent variable influences on the dependent variables. Finally, return on equity and financial leverage show positive significant association with the dependent variable of dividend payment, meanwhile corporate income tax gives negative significant relationship with dependent variable. Thus amongst the twelve variables selected from literature, these three variables are found to be affecting dividend policy matters in context of Vietnam. 5.2. Contribution The implications for shareholders from this study are that, with increase in the liability or the decrease in assets (financial leverage) they can expect a larger increase in dividend payout announcement for that year for being trusted the development of the 66 companies. Whereas corporate tax is consistent with the dividend payout implying that managers in Vietnam retain their dividend payouts, rather increase it even in the presence of corporate tax. Since dividend is also directly taxed, so the shareholder should be vigilant in his net proceeds, since a corporate tax is being levied on the income of the company yet dividends are being paid out to them which still have to be directly taxed (double taxation). For the managers, it’s a serious implication that the trend of dividend payout with increase in corporate taxation should be discouraged. Since increase in tax means decrease in net income. Managers should be focusing more on retained earnings rather than dividend payment with reference to corporate taxation since it’s of utmost significance in context of financial health of the firm under their management. 5.3. Recommendation As dividend is an unsolved puzzle, so there are implications for future research. The level impact of explanatory of the model in this research was given low through the econometric results implying further research for future. There are two further analyses being suggested to this paper. Firstly, in order to increase the generality of the experiment, a time chain should be done; for example, listed firms on the HOSE and even in HNX would be tested in a longer period which could be cover the crisis economy and the recovery period for comparing purpose. Secondly, questionnaires as well as interviews to directors of listed companies on the HOSE and HNX about factors driving dividend policy of their firms would be a more useful method for future research. 67 References Abdul Rehman, Haruto Takumi (2012) Determinants of dividend payout ratio: evidence from Karachi Stock Exchange (KSE). Journal of Contemporary Issues in Business Research. Vol. 1, Issue No.1, 20-27 Adedeji, A. (1998). Does the pecking order hypothesis explain the dividend payout ratios of firms in the UK? Journal of Business Finance and Accounting, 25(9–10), 1127– 1155. Ahmed, H. & Javid, A. (2009). Dynamics and Determinants of Dividend Policy in Pakistan Evidence from Karachi Stock Exchange. Non – Financial Listed Firms1450- 2887, Issue 25, -pp. 148-171 Aivazain, V., Booth, L. and Cleary, S. (2003). Dividend Policy And TheOrganization Of Capital Markets. Journal of Multinational Financial Management, vol. 13, No.2, pp. 101-121. Al- Kuwari, D. (2007). Determinants of the Dividend Payout Ratios of Companies Listed on Emerging Stock Exchanges. The Case of the Gulf Cooperation Council (GCC) Countries. PhD thesis, University of Wales-Cardiff, UK Al-Kuwari, D. (2009). Determinants of the Dividend Policy in Emerging Stock Exchanges:The Case of GCCCountries. Global Economy & Finance Journal, Vol. 2 No. 2 September 2009. pp. 38-63. Alli K, Khan A, Ramirez G. (1993). Determinants of dividend policy: a factorial analysis. Finance Rev, 28, 523-547 Allison, Paul D. (1999) Comparing Logit and Probit Coefficients Across Groups. Sociological Methods and Research 28:186-208. 68 Al-Najjar, B. and Hussainey, K. (2009). The Association between Dividend Payout and Outside Directorships. Journal of Applied Accounting Research, Vol. 10, No. 1, pp. 4-19. Amidu, Mohamed & Abor Joshua.(2006). Determinants of Dividend Payout Ratios in Ghana. The Journal of Risk Finance, 7 (2), 136-145. Ardestani, H. S, Zaleha, S. Abdul Rasid, R. B. & H Mehri (2013). Dividend Payout Policy, Investment Opportunity Set and Corporate Financing in the Industrial Products Sector of Malaysia. Journal of Applied Finance & Banking, vol. 3, no. 1. Asquith, P., & Mullins, D. W. (1983). The Impact of Initiating Dividend Payments on Shareholders' Wealth. The Journal of Business, 56 (1), 77-96. Bebczuk, R. (2005) Corporate Governance and Ownership: Measurement and Impact on Corporate Performance and Dividend Policies in Argentina. Center for Financial Stability, Working Paper. Bell, L. and T. Jenkinson (2002). New Evidence of the Impact of Dividend Taxation and on the Identity of the Marginal Investor. Journal of Finance 57, pp. 1321-1346. Bhattacharya, S. (1979). Imperfect Information, Dividend Policy, and "The Bird in the Hand" Fallacy. The Bell Journal of Economics, 10 (1), 259-270. Black, F. (1976). The dividend puzzle. Journal of Portfolio Management, 2 (2), 5-8 Black, F., & Scholes, M. (1974). The Effects of Dividend Yield and Dividend Policy on Common Stock Prices and Returns. Journal of Financial Economics, 1 (1), 1-22. Blundell, R., S. Bond and F. Windmeijer. (2000). Estimation in dynamic panel data models: improving on the performance of the standard GMM estimator. In B. 69 Baltagi (ed.), Nonstationary Panels, Panel Cointegration and Dynamic Panels, Elsevier Science Brigham, E. F., & Houston, J. F. (2011). Fundamentals of Financial Management. 7 ed. Mason: South-Western. Chang, R.P. and Rhee, S.G. (1990). The Impact of Personal Taxes on Corporate Dividend Policy and Capital Structure Decisions. Financial Management (Financial Management Association), 19(2), pp. 21-31. Chen, Z.,Cheung, Y., Stouraitis, A., & Wong, A., (2005). Ownership Concentration, Firm Performance, and Dividend Policy in Hong Kong. Pacific-Basin Finance Journal 13, 431-449. Easterbrook, F. H. (1984). Two Agency-Cost Explanations of Dividends. The American Economic Review, 74 (4), 650-659. Eddy, A. & Seifert, B. (1998), Firm size and dividend announcements. Journal of Financial Research, 11, 295-302. Eriotis, N. & Vasilou, D. (2003). Dividend Policy: An empirical Analysis of the Greek Market. International Business and Economics Research Journal, vol. 3, no.3, pp. 49-57. Farinha, J., (2003). Divided Policy. Corporate Governance and the Managerial Entrenchment Hypothesis: An Empirical Analysis. Journal of Business Finance & Accounting 30, 1173- 1209 Farrar, D., and L. Selwyn (1967) . Taxes, Corporate Financial Policy and Return to Investors. National Tax Journal, December, 1967: 444-454 70 Fenn, G.W., Liang, N. (2001) Corporate payout policy and managerial stock incentives. Journal of Financial Economics 60, 45-72 Franklin Allen & Antonio E. Bernardo & Ivo Welch (2000) A Theory of Dividends Based on Tax Clienteles. Journal of Finance, American Finance Association, vol. 55(6), pages 2499-2536, December. Freund, Rudolf J. and Ramon C. Littell (2000) SAS System for Regression (Third edition). Cary, NC: SAS Institute. Friend, I., & Puckett, M. (1964). Dividends and Stock Prices. The American Economic Review, 54 (5), 656-682. Gentry, William M., Deen Kemsley, and Christopher J. Mayer (2003). Dividend taxes and share prices: evidence from real estate investment trusts. Journal of Finance 58(1):261-282. Ghosh, C. and R. Woolridge (1988). An Analysis of Shareholder Reaction, Reaction to Dividend Cuts an Omissions. Journal of Financial Research, Winter 1988, pp.281294. Gill, A., Biger, N., & Tibrewala, R. (2010). Determinants of Dividend Payout Ratios: Evidence from United States. The Open Business Journal , III, 08-14. Gordon, M. J. (1959). Dividend, Earning, and Stock Prices. The Review of Economics and Statistics, Vol. 41, 1959, p: 99-105. Gordon, M. J. (1962). The Savings Investment and Valuation of a Corporation. The Review of Economics and Statistics, Vol. 44, 1962, p: 37-51. Gugler, K., & Yurtoglu, B., (2003). Corporate Governance and Dividend Pay-out Policy in Germany. European Economic Review 47, 731-758. 71 Higgins, R., (1972). The Corporate Dividend-Saving Decision , Journal of Financial and Quantitative Analysis 6, 1527-1541. HO, H (2003). Dividend policies in Australia and Japan. IAER, May, Vol. 9, No. 2. Holder, Mark E., Frederick W. Langrehr, and J. Lawrence Hexter, (1998). Dividend Policy Determinants: An Investigation of the Influences of Stakeholder Theory. Financial Management 27, 73-82. Jensen, M. and W. Meckling, (1992). Specific and general knowledge and organizational structure. In Contract Economics, eds. L. Werin and H. Wijkander, 251-274. Oxford: Blackwell. Jensen, M. C., & Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3 (4), 305-360. Kaleem, A. and Salahuddin, C. (2006), Impact of Dividend Announcement on Common Stock Prices at Lahore Stock Exchange (Pakistan), South Asian Journal of Management, Vol. 13 No.2 Kania, SL and Bacon, FW (2005) What factors motivate the corporate dividend decision. American Society of Business and Behavioral Sciences E-Journal,Vol. 1, No. 1. Kent Baker, H. (2009). Dividend and dividend policy. New Jersey: John Wiley & Sons. Keown, A. J., Martin, J. D., Petty, J. W., & Scott, D. F. (2007). Foundations of Finance: The Logic and Practice of Financial Management. 6 ed. London: Pearson Prentice Hall. Khan, M. Y., & Jain, P. K. (2008). Financial Management. New Dehli: Tata MsGrawHill Publishing Company 72 Kouki .Mondher and Guizani Moncef. (2009). Ownership Structure and Dividend Policy Evidence from the Tunisian Exchange market. Eurojournals Publishing, Inc. 2009. Kowalewski, O., Stetsyuk, I. and Talavera. (2007), Corporate Governance and Dividend Policy in Poland, Working Paper. German Institute for Economic Research. Kowalewski, o., stetsyuk, i. and talavera. (2007), Corporate Governance and Dividend Policy in Poland, Working Paper. German Institute for Economic Research. La Porta, R., F. Lopez-de-silanes, A. Schiefer and R. Vishny (1998) Agency problems and dividend policies around the world. Journal of Finance 55, 1-30. Lasfer, Ameziane M. (1995). Ex-Day Behavior: Tax or Short-Term Trading Effects. The Journal of Finance, 50(3), 875-897 Laub, P. M. (1976). On the informational content of dividends, Journal of Business. 49, pp. 73–80. Lease, Ronald C., Kose John, Avner Kalay, Uri Loewenstein, and Oded H. Sarig, (2000). Dividend Policy: Its Impact on Firm Value. Harvard Business School Press, Boston, Massachusttes. Li, K. and Zhao, X. (2008), Asymmetric Information and Dividend Policy. Financial Management, 37, 673-694. Lintner, John (1956). The Distribution of Income of Corporations Among Dividends, Retained Earnings, and Taxes,American Economic Review, 46, pp. 97-113. Lloyd, W. P., Jahera, J. S. & Page, D. E. (1985). Agency costs and dividend-payout ratios. Quarterly Journal of Business and Economics, Summer, 19-29. 73 Lonie, A.A., Abeyratna, G., Power, D.M. and Sinclair, C.D. (1996). The stock market reaction to dividend announcements: a study of complex market signals. Journal of Economic Studies, Vol. 23 No. 1, pp. 32 -52. Masulis, R.W. and Trueman, B. (1988). Corporate Investment & Dividend Decisions Under Differential Personal Taxation. Journal of Finance and Quantitative Analysis, 23, 369-386. Maury, C., Benjamin & Pajuste, A., (2002). Controlling Shareholders, Agency Problems, and Dividend Policy in Finland. In Working Paper. Stockholm School of Business: Stockholm School of Economics. McCabe, G., (1979). The Empirical Relationship between Investment and Financing: A New Look , Journal of Financial and Quantitative Analysis 14, 119-135. McCluskey, T. (2006). An Empirical Investigation of the Dividend Decison in Irish Companies. Unpublished PhD Thesis, School of Business, University of Dundee. Mehar, A. (2002). Corporate Governance And Dividend Policy. Mpra Paper,pp. 619623. Miller, M. H., & Rock, K. (1985). Dividend Policy under Asymmetric Information. The Journal of Finance, 40 (4), 1031-1051. Miller, Merton and Myron Scholes, (1978). Dividends and Taxes, Journal of Financial Economics, 6, 333-264. Miller, Merton H. (1986). Behavior Rationality in Finance : the case of dividends . Journal of Business, 59, S451-S468 Miller, Merton H. and Modigliani, F. (1961). Dividend Policy, Growth and the Valuation of Shares. Journal of Business, 34, 411-33. 74 Mohd, M., L. Perry, and J. Rimbey (1995). An Investigation of Dynamic Relationship Between Agency Theory and Dividend Policy. The Financial Review 30, 367-385. Mollah, S., K. Keasey, and H. Short (2002) The Influence of Agency Costs on Dividend Policy in an Emerging Market: Evidence from the Dhaka Stock Exchange, Working Paper Murphy , K. J. (1985). Corporate performance and managerial remuneration: An empirical analysis. Journal of Accounting and Economics, 7 (1-3), 11-42. Myers, M., & Bacon, F. (2004). The determinants of corporate dividend policy. Academy of Accounting and Financial Studies Journal, 8 (3), 17-28. Naceur, S., Goaied, M. and Belanes, A. (2006), On the Determinants and Dynamics of Dividend Policy, International Review of Finance, Vol. 6, No. 1-2, pp. 1- 23. Omran M., Pointon J., (2004). Dividend Policy, Trading Characteristics and Share Prices: Empirical evidence from Egyptian firms. International Journal of Theoretical and Applied Finance, 7(2),121-133 Pettit, R. (1972). Dividend announcemens, Security Performance, and Capital Market Efficiency. Journal of Finance, 27 (5), 993-1007. Pruitt, S.W. & Gitman, L.W. (1991). The interactions between the investment, financing, and dividend decisions of major US firms. Finance Review, 26: 409-30 Ramli, N. M. (2010). Ownership Structure and Dividend Policy: Evidence from Malaysian Companies. International Review of Business Research Papers, Vol. 6, No. 1, 170-180.Renneboog and Trojanowski (2005). Renneboog, L., Trojanowski, G.(2006). Control structures and payout policy. Managerial Finance 21, in preparation. 75 Royce Singleton Jr. and Bruce C. Straits. (1999). Approaches to Social Research. Oxford University Press. Rozeff , M. S. (1982). Growth, Beta and Agency Costs as Determinants of Dividend Payout Ratios. Journal of Financial Research, 5 (3), 249-259. Soter, Dennis, Eugene Brigham, and Paul Evanson. (1996). The Dividend Cut ‘Heard ’Round the World’: The Case of FPL. Journal of Applied Corporate Finance 9: 415. Tom S. Clark, Drew A. Linzer, March, (2012). Should I use fixed or random effects? Uddin, M. H. (2003). Effect of Dividend Announcement on Shareholders‟ Value: Evidence from Dhaka Stock Exchange. Global Corporate Governance Forum (Sponsored by World Bank), 13-16 December, Hyderabad, (INDIA) Venkatesh, P. C. (1989). The impact of dividend initiation on the information content of earnings announcements and returns volatility. Journal of Business, 62, pp. 175–197. Vogt, S. C. (1994). The cash flow/investment relationship: Evidence from U.S. manufacturing firms. Financial Management, Summer, 3-16. Watts, R. (1973). The information content of dividends. Journal of Business, 46, pp. 191– 211. Wooldridge, J.M. (2002). Econometric Analysis of Cross Section and Panel Data. Cambridge, MA: MIT Press Yin, R. K. (1994). Case study research: Design and methods (2nd ed.). Newbury Park, CA: Sage Publications. Zikmund, W. G. (1997). Business Research Method. Fifth ed. The Dryden press, Harcourt Brace College Publishers, Orlando, Florida. [...]... determine the dividend payout ratio in HOSE’s market  To analyze the impact of proposed determinants on dividend payout ratio policy 1.5 Research questions Thus, the research questions of this study are: - What is the specific relationship between the dividend payout ratio and selected factors for listed company in HOSE? - Whether the degree of significance determinants on payout ratio policy is homologous... consider when making dividend payout decisions Only a little literature, especially for dividend determinants in Vietnam has focused on empirical estimations 1.2 Rationale It is generally recognized that the particular choice of dividend payout by a business at a given point of time has important implications for a variety of considerations of significance to management, particularly those of future investments... main purpose of this paper research is to find out the association between dividend policy and financial leverage, investment, risk, corporate tax, liquidity, size, structure of asset, sale growth, cash flow, market- to-book value, ownership concentration, sale size, and return on equity of listed companies on Ho Chi Minh City Stock Exchange (HOSE) The sample consisted of 301 listed firms on HOSE in five... the dividend is the profit-sharing to the company’s investors, compensating them back part of the capital Likewise the strategic options, the dividend is considered as an easy way for investors to get profit from possessing stocks without trading them Normally, the board of directors gives the decision of how much should pay as dividend to the shareholders of a firm The matter of dividend payout ratio. .. determine the dividend policy for companies on HoChiMinh City stock market By studying dividend policy of these companies using elements model, it is hoped to create a basis for future study of dividend determinants of other industry in Vietnam too The subject can be extended by the analysis of other policies by stock dividend, buying back shares and evaluating the combined effects of all of these dividend. .. finance, dividend policy is paid much attention on Based on the business perspective, the dividend is a significant source of money to be settled outside For instance, in particular 1995, the listed companies on the New York Stock Exchange had paid up the amount up to USD 130 billion of dividends for shareholders In terms of research, the dividend policy is a matter of controversy The question of whether... Thematic objective is based on the theories and practical evidences of dividend policies around the world to reach the analysis and evaluation of the dividend policy of listed companies on the HoChiMinh City stock market, which can be representative 6 for Vietnam stock market Since then, a significance proposal is set up for building appropriate dividend policy for the listed companies in the future Therefore,... stock dividends are those made payment in the form of supplementary stock/ stocks of the issuing firm Commonly, stock dividends are issued in fraction to share owners To support and to have ideal measure of the factors analyzing dividend payout policy, the study consults both cash and stock dividend Several intellectual publications have pursuit to build up theoretical models that managers should consider... policy an unsolved puzzle Although a large scale amount of research regarding dividends has been conducted, there is no homogeneous response for the question: what are the determinants of the companies dividend payout ratios? There exist many reasons for 10 why companies should pay or not to pay dividends For example, the dividend payout is critical for investors because:  Dividends present certainty... determinants of dividend payout based on companies listed on HOSE 9 Chapter Two – Literature review The main purpose of this chapter is examining historically significant research studies that act as the basis for the proposed study With a view to implement the main purpose of this paper, the component problem statements are identified as follow: Determinants of dividend payout ratios: Evidence from .. .DETERMINANTS OF DIVIDEND PAYOUT RATIOS: EVIDENCE FROM LISTED COMPANIES ON HO CHI MINH STOCK MARKET In Partial Fulfillment of the Requirements of the Degree of MASTER OF BUSINESS ADMINISTRATION... the Determinants of dividend payout ratios: Evidence from listed companies on Ho Chi Minh Stock Market for the period of 2008-2012 would not have been possible without valuable contribution of. .. structure of asset, sale growth, cash flow, market- to-book value, ownership concentration, sale size, and return on equity of listed companies on Ho Chi Minh City Stock Exchange (HOSE) The sample consisted